The CMSIS DSP 5 library
Dependents: Nucleo-Heart-Rate ejercicioVrms2 PROYECTOFINAL ejercicioVrms ... more
functions/BasicMathFunctions/arm_shift_q7.c@3:4098b9d3d571, 2018-06-21 (annotated)
- Committer:
- xorjoep
- Date:
- Thu Jun 21 11:56:27 2018 +0000
- Revision:
- 3:4098b9d3d571
- Parent:
- 1:24714b45cd1b
headers is a folder not a library
Who changed what in which revision?
User | Revision | Line number | New contents of line |
---|---|---|---|
xorjoep | 1:24714b45cd1b | 1 | /* ---------------------------------------------------------------------- |
xorjoep | 1:24714b45cd1b | 2 | * Project: CMSIS DSP Library |
xorjoep | 1:24714b45cd1b | 3 | * Title: arm_shift_q7.c |
xorjoep | 1:24714b45cd1b | 4 | * Description: Processing function for the Q7 Shifting |
xorjoep | 1:24714b45cd1b | 5 | * |
xorjoep | 1:24714b45cd1b | 6 | * $Date: 27. January 2017 |
xorjoep | 1:24714b45cd1b | 7 | * $Revision: V.1.5.1 |
xorjoep | 1:24714b45cd1b | 8 | * |
xorjoep | 1:24714b45cd1b | 9 | * Target Processor: Cortex-M cores |
xorjoep | 1:24714b45cd1b | 10 | * -------------------------------------------------------------------- */ |
xorjoep | 1:24714b45cd1b | 11 | /* |
xorjoep | 1:24714b45cd1b | 12 | * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved. |
xorjoep | 1:24714b45cd1b | 13 | * |
xorjoep | 1:24714b45cd1b | 14 | * SPDX-License-Identifier: Apache-2.0 |
xorjoep | 1:24714b45cd1b | 15 | * |
xorjoep | 1:24714b45cd1b | 16 | * Licensed under the Apache License, Version 2.0 (the License); you may |
xorjoep | 1:24714b45cd1b | 17 | * not use this file except in compliance with the License. |
xorjoep | 1:24714b45cd1b | 18 | * You may obtain a copy of the License at |
xorjoep | 1:24714b45cd1b | 19 | * |
xorjoep | 1:24714b45cd1b | 20 | * www.apache.org/licenses/LICENSE-2.0 |
xorjoep | 1:24714b45cd1b | 21 | * |
xorjoep | 1:24714b45cd1b | 22 | * Unless required by applicable law or agreed to in writing, software |
xorjoep | 1:24714b45cd1b | 23 | * distributed under the License is distributed on an AS IS BASIS, WITHOUT |
xorjoep | 1:24714b45cd1b | 24 | * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
xorjoep | 1:24714b45cd1b | 25 | * See the License for the specific language governing permissions and |
xorjoep | 1:24714b45cd1b | 26 | * limitations under the License. |
xorjoep | 1:24714b45cd1b | 27 | */ |
xorjoep | 1:24714b45cd1b | 28 | |
xorjoep | 1:24714b45cd1b | 29 | #include "arm_math.h" |
xorjoep | 1:24714b45cd1b | 30 | |
xorjoep | 1:24714b45cd1b | 31 | /** |
xorjoep | 1:24714b45cd1b | 32 | * @ingroup groupMath |
xorjoep | 1:24714b45cd1b | 33 | */ |
xorjoep | 1:24714b45cd1b | 34 | |
xorjoep | 1:24714b45cd1b | 35 | /** |
xorjoep | 1:24714b45cd1b | 36 | * @addtogroup shift |
xorjoep | 1:24714b45cd1b | 37 | * @{ |
xorjoep | 1:24714b45cd1b | 38 | */ |
xorjoep | 1:24714b45cd1b | 39 | |
xorjoep | 1:24714b45cd1b | 40 | |
xorjoep | 1:24714b45cd1b | 41 | /** |
xorjoep | 1:24714b45cd1b | 42 | * @brief Shifts the elements of a Q7 vector a specified number of bits. |
xorjoep | 1:24714b45cd1b | 43 | * @param[in] *pSrc points to the input vector |
xorjoep | 1:24714b45cd1b | 44 | * @param[in] shiftBits number of bits to shift. A positive value shifts left; a negative value shifts right. |
xorjoep | 1:24714b45cd1b | 45 | * @param[out] *pDst points to the output vector |
xorjoep | 1:24714b45cd1b | 46 | * @param[in] blockSize number of samples in the vector |
xorjoep | 1:24714b45cd1b | 47 | * @return none. |
xorjoep | 1:24714b45cd1b | 48 | * |
xorjoep | 1:24714b45cd1b | 49 | * \par Conditions for optimum performance |
xorjoep | 1:24714b45cd1b | 50 | * Input and output buffers should be aligned by 32-bit |
xorjoep | 1:24714b45cd1b | 51 | * |
xorjoep | 1:24714b45cd1b | 52 | * |
xorjoep | 1:24714b45cd1b | 53 | * <b>Scaling and Overflow Behavior:</b> |
xorjoep | 1:24714b45cd1b | 54 | * \par |
xorjoep | 1:24714b45cd1b | 55 | * The function uses saturating arithmetic. |
xorjoep | 1:24714b45cd1b | 56 | * Results outside of the allowable Q7 range [0x8 0x7F] will be saturated. |
xorjoep | 1:24714b45cd1b | 57 | */ |
xorjoep | 1:24714b45cd1b | 58 | |
xorjoep | 1:24714b45cd1b | 59 | void arm_shift_q7( |
xorjoep | 1:24714b45cd1b | 60 | q7_t * pSrc, |
xorjoep | 1:24714b45cd1b | 61 | int8_t shiftBits, |
xorjoep | 1:24714b45cd1b | 62 | q7_t * pDst, |
xorjoep | 1:24714b45cd1b | 63 | uint32_t blockSize) |
xorjoep | 1:24714b45cd1b | 64 | { |
xorjoep | 1:24714b45cd1b | 65 | uint32_t blkCnt; /* loop counter */ |
xorjoep | 1:24714b45cd1b | 66 | uint8_t sign; /* Sign of shiftBits */ |
xorjoep | 1:24714b45cd1b | 67 | |
xorjoep | 1:24714b45cd1b | 68 | #if defined (ARM_MATH_DSP) |
xorjoep | 1:24714b45cd1b | 69 | |
xorjoep | 1:24714b45cd1b | 70 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
xorjoep | 1:24714b45cd1b | 71 | q7_t in1; /* Input value1 */ |
xorjoep | 1:24714b45cd1b | 72 | q7_t in2; /* Input value2 */ |
xorjoep | 1:24714b45cd1b | 73 | q7_t in3; /* Input value3 */ |
xorjoep | 1:24714b45cd1b | 74 | q7_t in4; /* Input value4 */ |
xorjoep | 1:24714b45cd1b | 75 | |
xorjoep | 1:24714b45cd1b | 76 | |
xorjoep | 1:24714b45cd1b | 77 | /*loop Unrolling */ |
xorjoep | 1:24714b45cd1b | 78 | blkCnt = blockSize >> 2U; |
xorjoep | 1:24714b45cd1b | 79 | |
xorjoep | 1:24714b45cd1b | 80 | /* Getting the sign of shiftBits */ |
xorjoep | 1:24714b45cd1b | 81 | sign = (shiftBits & 0x80); |
xorjoep | 1:24714b45cd1b | 82 | |
xorjoep | 1:24714b45cd1b | 83 | /* If the shift value is positive then do right shift else left shift */ |
xorjoep | 1:24714b45cd1b | 84 | if (sign == 0U) |
xorjoep | 1:24714b45cd1b | 85 | { |
xorjoep | 1:24714b45cd1b | 86 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
xorjoep | 1:24714b45cd1b | 87 | ** a second loop below computes the remaining 1 to 3 samples. */ |
xorjoep | 1:24714b45cd1b | 88 | while (blkCnt > 0U) |
xorjoep | 1:24714b45cd1b | 89 | { |
xorjoep | 1:24714b45cd1b | 90 | /* C = A << shiftBits */ |
xorjoep | 1:24714b45cd1b | 91 | /* Read 4 inputs */ |
xorjoep | 1:24714b45cd1b | 92 | in1 = *pSrc; |
xorjoep | 1:24714b45cd1b | 93 | in2 = *(pSrc + 1); |
xorjoep | 1:24714b45cd1b | 94 | in3 = *(pSrc + 2); |
xorjoep | 1:24714b45cd1b | 95 | in4 = *(pSrc + 3); |
xorjoep | 1:24714b45cd1b | 96 | |
xorjoep | 1:24714b45cd1b | 97 | /* Store the Shifted result in the destination buffer in single cycle by packing the outputs */ |
xorjoep | 1:24714b45cd1b | 98 | *__SIMD32(pDst)++ = __PACKq7(__SSAT((in1 << shiftBits), 8), |
xorjoep | 1:24714b45cd1b | 99 | __SSAT((in2 << shiftBits), 8), |
xorjoep | 1:24714b45cd1b | 100 | __SSAT((in3 << shiftBits), 8), |
xorjoep | 1:24714b45cd1b | 101 | __SSAT((in4 << shiftBits), 8)); |
xorjoep | 1:24714b45cd1b | 102 | /* Update source pointer to process next sampels */ |
xorjoep | 1:24714b45cd1b | 103 | pSrc += 4U; |
xorjoep | 1:24714b45cd1b | 104 | |
xorjoep | 1:24714b45cd1b | 105 | /* Decrement the loop counter */ |
xorjoep | 1:24714b45cd1b | 106 | blkCnt--; |
xorjoep | 1:24714b45cd1b | 107 | } |
xorjoep | 1:24714b45cd1b | 108 | |
xorjoep | 1:24714b45cd1b | 109 | /* If the blockSize is not a multiple of 4, compute any remaining output samples here. |
xorjoep | 1:24714b45cd1b | 110 | ** No loop unrolling is used. */ |
xorjoep | 1:24714b45cd1b | 111 | blkCnt = blockSize % 0x4U; |
xorjoep | 1:24714b45cd1b | 112 | |
xorjoep | 1:24714b45cd1b | 113 | while (blkCnt > 0U) |
xorjoep | 1:24714b45cd1b | 114 | { |
xorjoep | 1:24714b45cd1b | 115 | /* C = A << shiftBits */ |
xorjoep | 1:24714b45cd1b | 116 | /* Shift the input and then store the result in the destination buffer. */ |
xorjoep | 1:24714b45cd1b | 117 | *pDst++ = (q7_t) __SSAT((*pSrc++ << shiftBits), 8); |
xorjoep | 1:24714b45cd1b | 118 | |
xorjoep | 1:24714b45cd1b | 119 | /* Decrement the loop counter */ |
xorjoep | 1:24714b45cd1b | 120 | blkCnt--; |
xorjoep | 1:24714b45cd1b | 121 | } |
xorjoep | 1:24714b45cd1b | 122 | } |
xorjoep | 1:24714b45cd1b | 123 | else |
xorjoep | 1:24714b45cd1b | 124 | { |
xorjoep | 1:24714b45cd1b | 125 | shiftBits = -shiftBits; |
xorjoep | 1:24714b45cd1b | 126 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
xorjoep | 1:24714b45cd1b | 127 | ** a second loop below computes the remaining 1 to 3 samples. */ |
xorjoep | 1:24714b45cd1b | 128 | while (blkCnt > 0U) |
xorjoep | 1:24714b45cd1b | 129 | { |
xorjoep | 1:24714b45cd1b | 130 | /* C = A >> shiftBits */ |
xorjoep | 1:24714b45cd1b | 131 | /* Read 4 inputs */ |
xorjoep | 1:24714b45cd1b | 132 | in1 = *pSrc; |
xorjoep | 1:24714b45cd1b | 133 | in2 = *(pSrc + 1); |
xorjoep | 1:24714b45cd1b | 134 | in3 = *(pSrc + 2); |
xorjoep | 1:24714b45cd1b | 135 | in4 = *(pSrc + 3); |
xorjoep | 1:24714b45cd1b | 136 | |
xorjoep | 1:24714b45cd1b | 137 | /* Store the Shifted result in the destination buffer in single cycle by packing the outputs */ |
xorjoep | 1:24714b45cd1b | 138 | *__SIMD32(pDst)++ = __PACKq7((in1 >> shiftBits), (in2 >> shiftBits), |
xorjoep | 1:24714b45cd1b | 139 | (in3 >> shiftBits), (in4 >> shiftBits)); |
xorjoep | 1:24714b45cd1b | 140 | |
xorjoep | 1:24714b45cd1b | 141 | |
xorjoep | 1:24714b45cd1b | 142 | pSrc += 4U; |
xorjoep | 1:24714b45cd1b | 143 | |
xorjoep | 1:24714b45cd1b | 144 | /* Decrement the loop counter */ |
xorjoep | 1:24714b45cd1b | 145 | blkCnt--; |
xorjoep | 1:24714b45cd1b | 146 | } |
xorjoep | 1:24714b45cd1b | 147 | |
xorjoep | 1:24714b45cd1b | 148 | /* If the blockSize is not a multiple of 4, compute any remaining output samples here. |
xorjoep | 1:24714b45cd1b | 149 | ** No loop unrolling is used. */ |
xorjoep | 1:24714b45cd1b | 150 | blkCnt = blockSize % 0x4U; |
xorjoep | 1:24714b45cd1b | 151 | |
xorjoep | 1:24714b45cd1b | 152 | while (blkCnt > 0U) |
xorjoep | 1:24714b45cd1b | 153 | { |
xorjoep | 1:24714b45cd1b | 154 | /* C = A >> shiftBits */ |
xorjoep | 1:24714b45cd1b | 155 | /* Shift the input and then store the result in the destination buffer. */ |
xorjoep | 1:24714b45cd1b | 156 | in1 = *pSrc++; |
xorjoep | 1:24714b45cd1b | 157 | *pDst++ = (in1 >> shiftBits); |
xorjoep | 1:24714b45cd1b | 158 | |
xorjoep | 1:24714b45cd1b | 159 | /* Decrement the loop counter */ |
xorjoep | 1:24714b45cd1b | 160 | blkCnt--; |
xorjoep | 1:24714b45cd1b | 161 | } |
xorjoep | 1:24714b45cd1b | 162 | } |
xorjoep | 1:24714b45cd1b | 163 | |
xorjoep | 1:24714b45cd1b | 164 | #else |
xorjoep | 1:24714b45cd1b | 165 | |
xorjoep | 1:24714b45cd1b | 166 | /* Run the below code for Cortex-M0 */ |
xorjoep | 1:24714b45cd1b | 167 | |
xorjoep | 1:24714b45cd1b | 168 | /* Getting the sign of shiftBits */ |
xorjoep | 1:24714b45cd1b | 169 | sign = (shiftBits & 0x80); |
xorjoep | 1:24714b45cd1b | 170 | |
xorjoep | 1:24714b45cd1b | 171 | /* If the shift value is positive then do right shift else left shift */ |
xorjoep | 1:24714b45cd1b | 172 | if (sign == 0U) |
xorjoep | 1:24714b45cd1b | 173 | { |
xorjoep | 1:24714b45cd1b | 174 | /* Initialize blkCnt with number of samples */ |
xorjoep | 1:24714b45cd1b | 175 | blkCnt = blockSize; |
xorjoep | 1:24714b45cd1b | 176 | |
xorjoep | 1:24714b45cd1b | 177 | while (blkCnt > 0U) |
xorjoep | 1:24714b45cd1b | 178 | { |
xorjoep | 1:24714b45cd1b | 179 | /* C = A << shiftBits */ |
xorjoep | 1:24714b45cd1b | 180 | /* Shift the input and then store the result in the destination buffer. */ |
xorjoep | 1:24714b45cd1b | 181 | *pDst++ = (q7_t) __SSAT(((q15_t) * pSrc++ << shiftBits), 8); |
xorjoep | 1:24714b45cd1b | 182 | |
xorjoep | 1:24714b45cd1b | 183 | /* Decrement the loop counter */ |
xorjoep | 1:24714b45cd1b | 184 | blkCnt--; |
xorjoep | 1:24714b45cd1b | 185 | } |
xorjoep | 1:24714b45cd1b | 186 | } |
xorjoep | 1:24714b45cd1b | 187 | else |
xorjoep | 1:24714b45cd1b | 188 | { |
xorjoep | 1:24714b45cd1b | 189 | /* Initialize blkCnt with number of samples */ |
xorjoep | 1:24714b45cd1b | 190 | blkCnt = blockSize; |
xorjoep | 1:24714b45cd1b | 191 | |
xorjoep | 1:24714b45cd1b | 192 | while (blkCnt > 0U) |
xorjoep | 1:24714b45cd1b | 193 | { |
xorjoep | 1:24714b45cd1b | 194 | /* C = A >> shiftBits */ |
xorjoep | 1:24714b45cd1b | 195 | /* Shift the input and then store the result in the destination buffer. */ |
xorjoep | 1:24714b45cd1b | 196 | *pDst++ = (*pSrc++ >> -shiftBits); |
xorjoep | 1:24714b45cd1b | 197 | |
xorjoep | 1:24714b45cd1b | 198 | /* Decrement the loop counter */ |
xorjoep | 1:24714b45cd1b | 199 | blkCnt--; |
xorjoep | 1:24714b45cd1b | 200 | } |
xorjoep | 1:24714b45cd1b | 201 | } |
xorjoep | 1:24714b45cd1b | 202 | |
xorjoep | 1:24714b45cd1b | 203 | #endif /* #if defined (ARM_MATH_DSP) */ |
xorjoep | 1:24714b45cd1b | 204 | } |
xorjoep | 1:24714b45cd1b | 205 | |
xorjoep | 1:24714b45cd1b | 206 | /** |
xorjoep | 1:24714b45cd1b | 207 | * @} end of shift group |
xorjoep | 1:24714b45cd1b | 208 | */ |