mbed-os 6.10 versione
cmsis_dsp/ComplexMathFunctions/arm_cmplx_mult_cmplx_q31.c@2:da51fb522205, 2013-05-30 (annotated)
- Committer:
- emilmont
- Date:
- Thu May 30 17:10:11 2013 +0100
- Revision:
- 2:da51fb522205
- Parent:
- 1:fdd22bb7aa52
- Child:
- 3:7a284390b0ce
Keep "cmsis-dsp" module in synch with its source
Who changed what in which revision?
User | Revision | Line number | New contents of line |
---|---|---|---|
emilmont | 1:fdd22bb7aa52 | 1 | /* ---------------------------------------------------------------------- |
emilmont | 1:fdd22bb7aa52 | 2 | * Copyright (C) 2010 ARM Limited. All rights reserved. |
emilmont | 1:fdd22bb7aa52 | 3 | * |
emilmont | 1:fdd22bb7aa52 | 4 | * $Date: 15. February 2012 |
emilmont | 2:da51fb522205 | 5 | * $Revision: V1.1.0 |
emilmont | 1:fdd22bb7aa52 | 6 | * |
emilmont | 2:da51fb522205 | 7 | * Project: CMSIS DSP Library |
emilmont | 2:da51fb522205 | 8 | * Title: arm_cmplx_mult_cmplx_q31.c |
emilmont | 1:fdd22bb7aa52 | 9 | * |
emilmont | 2:da51fb522205 | 10 | * Description: Q31 complex-by-complex multiplication |
emilmont | 1:fdd22bb7aa52 | 11 | * |
emilmont | 1:fdd22bb7aa52 | 12 | * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0 |
emilmont | 1:fdd22bb7aa52 | 13 | * |
emilmont | 1:fdd22bb7aa52 | 14 | * Version 1.1.0 2012/02/15 |
emilmont | 1:fdd22bb7aa52 | 15 | * Updated with more optimizations, bug fixes and minor API changes. |
emilmont | 1:fdd22bb7aa52 | 16 | * |
emilmont | 1:fdd22bb7aa52 | 17 | * Version 1.0.10 2011/7/15 |
emilmont | 1:fdd22bb7aa52 | 18 | * Big Endian support added and Merged M0 and M3/M4 Source code. |
emilmont | 1:fdd22bb7aa52 | 19 | * |
emilmont | 1:fdd22bb7aa52 | 20 | * Version 1.0.3 2010/11/29 |
emilmont | 1:fdd22bb7aa52 | 21 | * Re-organized the CMSIS folders and updated documentation. |
emilmont | 1:fdd22bb7aa52 | 22 | * |
emilmont | 1:fdd22bb7aa52 | 23 | * Version 1.0.2 2010/11/11 |
emilmont | 1:fdd22bb7aa52 | 24 | * Documentation updated. |
emilmont | 1:fdd22bb7aa52 | 25 | * |
emilmont | 1:fdd22bb7aa52 | 26 | * Version 1.0.1 2010/10/05 |
emilmont | 1:fdd22bb7aa52 | 27 | * Production release and review comments incorporated. |
emilmont | 1:fdd22bb7aa52 | 28 | * |
emilmont | 1:fdd22bb7aa52 | 29 | * Version 1.0.0 2010/09/20 |
emilmont | 1:fdd22bb7aa52 | 30 | * Production release and review comments incorporated. |
emilmont | 1:fdd22bb7aa52 | 31 | * -------------------------------------------------------------------- */ |
emilmont | 1:fdd22bb7aa52 | 32 | |
emilmont | 1:fdd22bb7aa52 | 33 | #include "arm_math.h" |
emilmont | 1:fdd22bb7aa52 | 34 | |
emilmont | 1:fdd22bb7aa52 | 35 | /** |
emilmont | 1:fdd22bb7aa52 | 36 | * @ingroup groupCmplxMath |
emilmont | 1:fdd22bb7aa52 | 37 | */ |
emilmont | 1:fdd22bb7aa52 | 38 | |
emilmont | 1:fdd22bb7aa52 | 39 | /** |
emilmont | 1:fdd22bb7aa52 | 40 | * @addtogroup CmplxByCmplxMult |
emilmont | 1:fdd22bb7aa52 | 41 | * @{ |
emilmont | 1:fdd22bb7aa52 | 42 | */ |
emilmont | 1:fdd22bb7aa52 | 43 | |
emilmont | 1:fdd22bb7aa52 | 44 | |
emilmont | 1:fdd22bb7aa52 | 45 | /** |
emilmont | 1:fdd22bb7aa52 | 46 | * @brief Q31 complex-by-complex multiplication |
emilmont | 1:fdd22bb7aa52 | 47 | * @param[in] *pSrcA points to the first input vector |
emilmont | 1:fdd22bb7aa52 | 48 | * @param[in] *pSrcB points to the second input vector |
emilmont | 1:fdd22bb7aa52 | 49 | * @param[out] *pDst points to the output vector |
emilmont | 1:fdd22bb7aa52 | 50 | * @param[in] numSamples number of complex samples in each vector |
emilmont | 1:fdd22bb7aa52 | 51 | * @return none. |
emilmont | 1:fdd22bb7aa52 | 52 | * |
emilmont | 1:fdd22bb7aa52 | 53 | * <b>Scaling and Overflow Behavior:</b> |
emilmont | 1:fdd22bb7aa52 | 54 | * \par |
emilmont | 1:fdd22bb7aa52 | 55 | * The function implements 1.31 by 1.31 multiplications and finally output is converted into 3.29 format. |
emilmont | 1:fdd22bb7aa52 | 56 | * Input down scaling is not required. |
emilmont | 1:fdd22bb7aa52 | 57 | */ |
emilmont | 1:fdd22bb7aa52 | 58 | |
emilmont | 1:fdd22bb7aa52 | 59 | void arm_cmplx_mult_cmplx_q31( |
emilmont | 1:fdd22bb7aa52 | 60 | q31_t * pSrcA, |
emilmont | 1:fdd22bb7aa52 | 61 | q31_t * pSrcB, |
emilmont | 1:fdd22bb7aa52 | 62 | q31_t * pDst, |
emilmont | 1:fdd22bb7aa52 | 63 | uint32_t numSamples) |
emilmont | 1:fdd22bb7aa52 | 64 | { |
emilmont | 1:fdd22bb7aa52 | 65 | q31_t a, b, c, d; /* Temporary variables to store real and imaginary values */ |
emilmont | 1:fdd22bb7aa52 | 66 | uint32_t blkCnt; /* loop counters */ |
emilmont | 1:fdd22bb7aa52 | 67 | q31_t mul1, mul2, mul3, mul4; |
emilmont | 1:fdd22bb7aa52 | 68 | q31_t out1, out2; |
emilmont | 1:fdd22bb7aa52 | 69 | |
emilmont | 1:fdd22bb7aa52 | 70 | #ifndef ARM_MATH_CM0 |
emilmont | 1:fdd22bb7aa52 | 71 | |
emilmont | 1:fdd22bb7aa52 | 72 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
emilmont | 1:fdd22bb7aa52 | 73 | |
emilmont | 1:fdd22bb7aa52 | 74 | /* loop Unrolling */ |
emilmont | 1:fdd22bb7aa52 | 75 | blkCnt = numSamples >> 2u; |
emilmont | 1:fdd22bb7aa52 | 76 | |
emilmont | 1:fdd22bb7aa52 | 77 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
emilmont | 1:fdd22bb7aa52 | 78 | ** a second loop below computes the remaining 1 to 3 samples. */ |
emilmont | 1:fdd22bb7aa52 | 79 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 80 | { |
emilmont | 1:fdd22bb7aa52 | 81 | /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */ |
emilmont | 1:fdd22bb7aa52 | 82 | /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */ |
emilmont | 1:fdd22bb7aa52 | 83 | a = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 84 | b = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 85 | c = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 86 | d = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 87 | |
emilmont | 1:fdd22bb7aa52 | 88 | mul1 = (q31_t) (((q63_t) a * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 89 | mul2 = (q31_t) (((q63_t) b * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 90 | mul3 = (q31_t) (((q63_t) a * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 91 | mul4 = (q31_t) (((q63_t) b * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 92 | |
emilmont | 1:fdd22bb7aa52 | 93 | mul1 = (mul1 >> 1); |
emilmont | 1:fdd22bb7aa52 | 94 | mul2 = (mul2 >> 1); |
emilmont | 1:fdd22bb7aa52 | 95 | mul3 = (mul3 >> 1); |
emilmont | 1:fdd22bb7aa52 | 96 | mul4 = (mul4 >> 1); |
emilmont | 1:fdd22bb7aa52 | 97 | |
emilmont | 1:fdd22bb7aa52 | 98 | out1 = mul1 - mul2; |
emilmont | 1:fdd22bb7aa52 | 99 | out2 = mul3 + mul4; |
emilmont | 1:fdd22bb7aa52 | 100 | |
emilmont | 1:fdd22bb7aa52 | 101 | /* store the real result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 102 | *pDst++ = out1; |
emilmont | 1:fdd22bb7aa52 | 103 | /* store the imag result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 104 | *pDst++ = out2; |
emilmont | 1:fdd22bb7aa52 | 105 | |
emilmont | 1:fdd22bb7aa52 | 106 | a = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 107 | b = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 108 | c = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 109 | d = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 110 | |
emilmont | 1:fdd22bb7aa52 | 111 | mul1 = (q31_t) (((q63_t) a * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 112 | mul2 = (q31_t) (((q63_t) b * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 113 | mul3 = (q31_t) (((q63_t) a * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 114 | mul4 = (q31_t) (((q63_t) b * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 115 | |
emilmont | 1:fdd22bb7aa52 | 116 | mul1 = (mul1 >> 1); |
emilmont | 1:fdd22bb7aa52 | 117 | mul2 = (mul2 >> 1); |
emilmont | 1:fdd22bb7aa52 | 118 | mul3 = (mul3 >> 1); |
emilmont | 1:fdd22bb7aa52 | 119 | mul4 = (mul4 >> 1); |
emilmont | 1:fdd22bb7aa52 | 120 | |
emilmont | 1:fdd22bb7aa52 | 121 | out1 = mul1 - mul2; |
emilmont | 1:fdd22bb7aa52 | 122 | out2 = mul3 + mul4; |
emilmont | 1:fdd22bb7aa52 | 123 | |
emilmont | 1:fdd22bb7aa52 | 124 | /* store the real result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 125 | *pDst++ = out1; |
emilmont | 1:fdd22bb7aa52 | 126 | /* store the imag result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 127 | *pDst++ = out2; |
emilmont | 1:fdd22bb7aa52 | 128 | |
emilmont | 1:fdd22bb7aa52 | 129 | a = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 130 | b = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 131 | c = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 132 | d = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 133 | |
emilmont | 1:fdd22bb7aa52 | 134 | mul1 = (q31_t) (((q63_t) a * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 135 | mul2 = (q31_t) (((q63_t) b * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 136 | mul3 = (q31_t) (((q63_t) a * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 137 | mul4 = (q31_t) (((q63_t) b * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 138 | |
emilmont | 1:fdd22bb7aa52 | 139 | mul1 = (mul1 >> 1); |
emilmont | 1:fdd22bb7aa52 | 140 | mul2 = (mul2 >> 1); |
emilmont | 1:fdd22bb7aa52 | 141 | mul3 = (mul3 >> 1); |
emilmont | 1:fdd22bb7aa52 | 142 | mul4 = (mul4 >> 1); |
emilmont | 1:fdd22bb7aa52 | 143 | |
emilmont | 1:fdd22bb7aa52 | 144 | out1 = mul1 - mul2; |
emilmont | 1:fdd22bb7aa52 | 145 | out2 = mul3 + mul4; |
emilmont | 1:fdd22bb7aa52 | 146 | |
emilmont | 1:fdd22bb7aa52 | 147 | /* store the real result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 148 | *pDst++ = out1; |
emilmont | 1:fdd22bb7aa52 | 149 | /* store the imag result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 150 | *pDst++ = out2; |
emilmont | 1:fdd22bb7aa52 | 151 | |
emilmont | 1:fdd22bb7aa52 | 152 | a = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 153 | b = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 154 | c = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 155 | d = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 156 | |
emilmont | 1:fdd22bb7aa52 | 157 | mul1 = (q31_t) (((q63_t) a * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 158 | mul2 = (q31_t) (((q63_t) b * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 159 | mul3 = (q31_t) (((q63_t) a * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 160 | mul4 = (q31_t) (((q63_t) b * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 161 | |
emilmont | 1:fdd22bb7aa52 | 162 | mul1 = (mul1 >> 1); |
emilmont | 1:fdd22bb7aa52 | 163 | mul2 = (mul2 >> 1); |
emilmont | 1:fdd22bb7aa52 | 164 | mul3 = (mul3 >> 1); |
emilmont | 1:fdd22bb7aa52 | 165 | mul4 = (mul4 >> 1); |
emilmont | 1:fdd22bb7aa52 | 166 | |
emilmont | 1:fdd22bb7aa52 | 167 | out1 = mul1 - mul2; |
emilmont | 1:fdd22bb7aa52 | 168 | out2 = mul3 + mul4; |
emilmont | 1:fdd22bb7aa52 | 169 | |
emilmont | 1:fdd22bb7aa52 | 170 | /* store the real result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 171 | *pDst++ = out1; |
emilmont | 1:fdd22bb7aa52 | 172 | /* store the imag result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 173 | *pDst++ = out2; |
emilmont | 1:fdd22bb7aa52 | 174 | |
emilmont | 1:fdd22bb7aa52 | 175 | /* Decrement the blockSize loop counter */ |
emilmont | 1:fdd22bb7aa52 | 176 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 177 | } |
emilmont | 1:fdd22bb7aa52 | 178 | |
emilmont | 1:fdd22bb7aa52 | 179 | /* If the blockSize is not a multiple of 4, compute any remaining output samples here. |
emilmont | 1:fdd22bb7aa52 | 180 | ** No loop unrolling is used. */ |
emilmont | 1:fdd22bb7aa52 | 181 | blkCnt = numSamples % 0x4u; |
emilmont | 1:fdd22bb7aa52 | 182 | |
emilmont | 1:fdd22bb7aa52 | 183 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 184 | { |
emilmont | 1:fdd22bb7aa52 | 185 | /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */ |
emilmont | 1:fdd22bb7aa52 | 186 | /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */ |
emilmont | 1:fdd22bb7aa52 | 187 | a = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 188 | b = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 189 | c = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 190 | d = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 191 | |
emilmont | 1:fdd22bb7aa52 | 192 | mul1 = (q31_t) (((q63_t) a * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 193 | mul2 = (q31_t) (((q63_t) b * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 194 | mul3 = (q31_t) (((q63_t) a * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 195 | mul4 = (q31_t) (((q63_t) b * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 196 | |
emilmont | 1:fdd22bb7aa52 | 197 | mul1 = (mul1 >> 1); |
emilmont | 1:fdd22bb7aa52 | 198 | mul2 = (mul2 >> 1); |
emilmont | 1:fdd22bb7aa52 | 199 | mul3 = (mul3 >> 1); |
emilmont | 1:fdd22bb7aa52 | 200 | mul4 = (mul4 >> 1); |
emilmont | 1:fdd22bb7aa52 | 201 | |
emilmont | 1:fdd22bb7aa52 | 202 | out1 = mul1 - mul2; |
emilmont | 1:fdd22bb7aa52 | 203 | out2 = mul3 + mul4; |
emilmont | 1:fdd22bb7aa52 | 204 | |
emilmont | 1:fdd22bb7aa52 | 205 | /* store the real result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 206 | *pDst++ = out1; |
emilmont | 1:fdd22bb7aa52 | 207 | /* store the imag result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 208 | *pDst++ = out2; |
emilmont | 1:fdd22bb7aa52 | 209 | |
emilmont | 1:fdd22bb7aa52 | 210 | /* Decrement the blockSize loop counter */ |
emilmont | 1:fdd22bb7aa52 | 211 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 212 | } |
emilmont | 1:fdd22bb7aa52 | 213 | |
emilmont | 1:fdd22bb7aa52 | 214 | #else |
emilmont | 1:fdd22bb7aa52 | 215 | |
emilmont | 1:fdd22bb7aa52 | 216 | /* Run the below code for Cortex-M0 */ |
emilmont | 1:fdd22bb7aa52 | 217 | |
emilmont | 1:fdd22bb7aa52 | 218 | /* loop Unrolling */ |
emilmont | 1:fdd22bb7aa52 | 219 | blkCnt = numSamples >> 1u; |
emilmont | 1:fdd22bb7aa52 | 220 | |
emilmont | 1:fdd22bb7aa52 | 221 | /* First part of the processing with loop unrolling. Compute 2 outputs at a time. |
emilmont | 1:fdd22bb7aa52 | 222 | ** a second loop below computes the remaining 1 sample. */ |
emilmont | 1:fdd22bb7aa52 | 223 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 224 | { |
emilmont | 1:fdd22bb7aa52 | 225 | /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */ |
emilmont | 1:fdd22bb7aa52 | 226 | /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */ |
emilmont | 1:fdd22bb7aa52 | 227 | a = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 228 | b = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 229 | c = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 230 | d = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 231 | |
emilmont | 1:fdd22bb7aa52 | 232 | mul1 = (q31_t) (((q63_t) a * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 233 | mul2 = (q31_t) (((q63_t) b * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 234 | mul3 = (q31_t) (((q63_t) a * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 235 | mul4 = (q31_t) (((q63_t) b * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 236 | |
emilmont | 1:fdd22bb7aa52 | 237 | mul1 = (mul1 >> 1); |
emilmont | 1:fdd22bb7aa52 | 238 | mul2 = (mul2 >> 1); |
emilmont | 1:fdd22bb7aa52 | 239 | mul3 = (mul3 >> 1); |
emilmont | 1:fdd22bb7aa52 | 240 | mul4 = (mul4 >> 1); |
emilmont | 1:fdd22bb7aa52 | 241 | |
emilmont | 1:fdd22bb7aa52 | 242 | out1 = mul1 - mul2; |
emilmont | 1:fdd22bb7aa52 | 243 | out2 = mul3 + mul4; |
emilmont | 1:fdd22bb7aa52 | 244 | |
emilmont | 1:fdd22bb7aa52 | 245 | /* store the real result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 246 | *pDst++ = out1; |
emilmont | 1:fdd22bb7aa52 | 247 | /* store the imag result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 248 | *pDst++ = out2; |
emilmont | 1:fdd22bb7aa52 | 249 | |
emilmont | 1:fdd22bb7aa52 | 250 | a = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 251 | b = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 252 | c = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 253 | d = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 254 | |
emilmont | 1:fdd22bb7aa52 | 255 | mul1 = (q31_t) (((q63_t) a * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 256 | mul2 = (q31_t) (((q63_t) b * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 257 | mul3 = (q31_t) (((q63_t) a * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 258 | mul4 = (q31_t) (((q63_t) b * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 259 | |
emilmont | 1:fdd22bb7aa52 | 260 | mul1 = (mul1 >> 1); |
emilmont | 1:fdd22bb7aa52 | 261 | mul2 = (mul2 >> 1); |
emilmont | 1:fdd22bb7aa52 | 262 | mul3 = (mul3 >> 1); |
emilmont | 1:fdd22bb7aa52 | 263 | mul4 = (mul4 >> 1); |
emilmont | 1:fdd22bb7aa52 | 264 | |
emilmont | 1:fdd22bb7aa52 | 265 | out1 = mul1 - mul2; |
emilmont | 1:fdd22bb7aa52 | 266 | out2 = mul3 + mul4; |
emilmont | 1:fdd22bb7aa52 | 267 | |
emilmont | 1:fdd22bb7aa52 | 268 | /* store the real result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 269 | *pDst++ = out1; |
emilmont | 1:fdd22bb7aa52 | 270 | /* store the imag result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 271 | *pDst++ = out2; |
emilmont | 1:fdd22bb7aa52 | 272 | |
emilmont | 1:fdd22bb7aa52 | 273 | /* Decrement the blockSize loop counter */ |
emilmont | 1:fdd22bb7aa52 | 274 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 275 | } |
emilmont | 1:fdd22bb7aa52 | 276 | |
emilmont | 1:fdd22bb7aa52 | 277 | /* If the blockSize is not a multiple of 2, compute any remaining output samples here. |
emilmont | 1:fdd22bb7aa52 | 278 | ** No loop unrolling is used. */ |
emilmont | 1:fdd22bb7aa52 | 279 | blkCnt = numSamples % 0x2u; |
emilmont | 1:fdd22bb7aa52 | 280 | |
emilmont | 1:fdd22bb7aa52 | 281 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 282 | { |
emilmont | 1:fdd22bb7aa52 | 283 | /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */ |
emilmont | 1:fdd22bb7aa52 | 284 | /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */ |
emilmont | 1:fdd22bb7aa52 | 285 | a = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 286 | b = *pSrcA++; |
emilmont | 1:fdd22bb7aa52 | 287 | c = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 288 | d = *pSrcB++; |
emilmont | 1:fdd22bb7aa52 | 289 | |
emilmont | 1:fdd22bb7aa52 | 290 | mul1 = (q31_t) (((q63_t) a * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 291 | mul2 = (q31_t) (((q63_t) b * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 292 | mul3 = (q31_t) (((q63_t) a * d) >> 32); |
emilmont | 1:fdd22bb7aa52 | 293 | mul4 = (q31_t) (((q63_t) b * c) >> 32); |
emilmont | 1:fdd22bb7aa52 | 294 | |
emilmont | 1:fdd22bb7aa52 | 295 | mul1 = (mul1 >> 1); |
emilmont | 1:fdd22bb7aa52 | 296 | mul2 = (mul2 >> 1); |
emilmont | 1:fdd22bb7aa52 | 297 | mul3 = (mul3 >> 1); |
emilmont | 1:fdd22bb7aa52 | 298 | mul4 = (mul4 >> 1); |
emilmont | 1:fdd22bb7aa52 | 299 | |
emilmont | 1:fdd22bb7aa52 | 300 | out1 = mul1 - mul2; |
emilmont | 1:fdd22bb7aa52 | 301 | out2 = mul3 + mul4; |
emilmont | 1:fdd22bb7aa52 | 302 | |
emilmont | 1:fdd22bb7aa52 | 303 | /* store the real result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 304 | *pDst++ = out1; |
emilmont | 1:fdd22bb7aa52 | 305 | /* store the imag result in 3.29 format in the destination buffer. */ |
emilmont | 1:fdd22bb7aa52 | 306 | *pDst++ = out2; |
emilmont | 1:fdd22bb7aa52 | 307 | |
emilmont | 1:fdd22bb7aa52 | 308 | /* Decrement the blockSize loop counter */ |
emilmont | 1:fdd22bb7aa52 | 309 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 310 | } |
emilmont | 1:fdd22bb7aa52 | 311 | |
emilmont | 1:fdd22bb7aa52 | 312 | #endif /* #ifndef ARM_MATH_CM0 */ |
emilmont | 1:fdd22bb7aa52 | 313 | |
emilmont | 1:fdd22bb7aa52 | 314 | } |
emilmont | 1:fdd22bb7aa52 | 315 | |
emilmont | 1:fdd22bb7aa52 | 316 | /** |
emilmont | 1:fdd22bb7aa52 | 317 | * @} end of CmplxByCmplxMult group |
emilmont | 1:fdd22bb7aa52 | 318 | */ |