mbed official / mbed-dsp

CMSIS DSP library

Dependents:   performance_timer Surfboard_ gps2rtty Capstone ... more

Legacy Warning

This is an mbed 2 library. To learn more about mbed OS 5, visit the docs.

cmsis_dsp/FilteringFunctions/arm_fir_sparse_q31.c@1:fdd22bb7aa52, 2012-11-28 (annotated)

Committer:
emilmont
Date:
Wed Nov 28 12:30:09 2012 +0000
Revision:
1:fdd22bb7aa52
Child:
2:da51fb522205
DSP library code

Who changed what in which revision?

UserRevisionLine numberNew 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 1:fdd22bb7aa52 5 * $Revision: V1.1.0
emilmont 1:fdd22bb7aa52 6 *
emilmont 1:fdd22bb7aa52 7 * Project: CMSIS DSP Library
emilmont 1:fdd22bb7aa52 8 * Title: arm_fir_sparse_q31.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 1:fdd22bb7aa52 10 * Description: Q31 sparse FIR filter processing function.
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 * Version 0.0.7 2010/06/10
emilmont 1:fdd22bb7aa52 33 * Misra-C changes done
emilmont 1:fdd22bb7aa52 34 * ------------------------------------------------------------------- */
emilmont 1:fdd22bb7aa52 35 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 36
emilmont 1:fdd22bb7aa52 37
emilmont 1:fdd22bb7aa52 38 /**
emilmont 1:fdd22bb7aa52 39 * @addtogroup FIR_Sparse
emilmont 1:fdd22bb7aa52 40 * @{
emilmont 1:fdd22bb7aa52 41 */
emilmont 1:fdd22bb7aa52 42
emilmont 1:fdd22bb7aa52 43 /**
emilmont 1:fdd22bb7aa52 44 * @brief Processing function for the Q31 sparse FIR filter.
emilmont 1:fdd22bb7aa52 45 * @param[in] *S points to an instance of the Q31 sparse FIR structure.
emilmont 1:fdd22bb7aa52 46 * @param[in] *pSrc points to the block of input data.
emilmont 1:fdd22bb7aa52 47 * @param[out] *pDst points to the block of output data
emilmont 1:fdd22bb7aa52 48 * @param[in] *pScratchIn points to a temporary buffer of size blockSize.
emilmont 1:fdd22bb7aa52 49 * @param[in] blockSize number of input samples to process per call.
emilmont 1:fdd22bb7aa52 50 * @return none.
emilmont 1:fdd22bb7aa52 51 *
emilmont 1:fdd22bb7aa52 52 * <b>Scaling and Overflow Behavior:</b>
emilmont 1:fdd22bb7aa52 53 * \par
emilmont 1:fdd22bb7aa52 54 * The function is implemented using an internal 32-bit accumulator.
emilmont 1:fdd22bb7aa52 55 * The 1.31 x 1.31 multiplications are truncated to 2.30 format.
emilmont 1:fdd22bb7aa52 56 * This leads to loss of precision on the intermediate multiplications and provides only a single guard bit.
emilmont 1:fdd22bb7aa52 57 * If the accumulator result overflows, it wraps around rather than saturate.
emilmont 1:fdd22bb7aa52 58 * In order to avoid overflows the input signal or coefficients must be scaled down by log2(numTaps) bits.
emilmont 1:fdd22bb7aa52 59 */
emilmont 1:fdd22bb7aa52 60
emilmont 1:fdd22bb7aa52 61 void arm_fir_sparse_q31(
emilmont 1:fdd22bb7aa52 62 arm_fir_sparse_instance_q31 * S,
emilmont 1:fdd22bb7aa52 63 q31_t * pSrc,
emilmont 1:fdd22bb7aa52 64 q31_t * pDst,
emilmont 1:fdd22bb7aa52 65 q31_t * pScratchIn,
emilmont 1:fdd22bb7aa52 66 uint32_t blockSize)
emilmont 1:fdd22bb7aa52 67 {
emilmont 1:fdd22bb7aa52 68
emilmont 1:fdd22bb7aa52 69 q31_t *pState = S->pState; /* State pointer */
emilmont 1:fdd22bb7aa52 70 q31_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emilmont 1:fdd22bb7aa52 71 q31_t *px; /* Scratch buffer pointer */
emilmont 1:fdd22bb7aa52 72 q31_t *py = pState; /* Temporary pointers for state buffer */
emilmont 1:fdd22bb7aa52 73 q31_t *pb = pScratchIn; /* Temporary pointers for scratch buffer */
emilmont 1:fdd22bb7aa52 74 q31_t *pOut; /* Destination pointer */
emilmont 1:fdd22bb7aa52 75 q63_t out; /* Temporary output variable */
emilmont 1:fdd22bb7aa52 76 int32_t *pTapDelay = S->pTapDelay; /* Pointer to the array containing offset of the non-zero tap values. */
emilmont 1:fdd22bb7aa52 77 uint32_t delaySize = S->maxDelay + blockSize; /* state length */
emilmont 1:fdd22bb7aa52 78 uint16_t numTaps = S->numTaps; /* Filter order */
emilmont 1:fdd22bb7aa52 79 int32_t readIndex; /* Read index of the state buffer */
emilmont 1:fdd22bb7aa52 80 uint32_t tapCnt, blkCnt; /* loop counters */
emilmont 1:fdd22bb7aa52 81 q31_t coeff = *pCoeffs++; /* Read the first coefficient value */
emilmont 1:fdd22bb7aa52 82 q31_t in;
emilmont 1:fdd22bb7aa52 83
emilmont 1:fdd22bb7aa52 84
emilmont 1:fdd22bb7aa52 85 /* BlockSize of Input samples are copied into the state buffer */
emilmont 1:fdd22bb7aa52 86 /* StateIndex points to the starting position to write in the state buffer */
emilmont 1:fdd22bb7aa52 87 arm_circularWrite_f32((int32_t *) py, delaySize, &S->stateIndex, 1,
emilmont 1:fdd22bb7aa52 88 (int32_t *) pSrc, 1, blockSize);
emilmont 1:fdd22bb7aa52 89
emilmont 1:fdd22bb7aa52 90 /* Read Index, from where the state buffer should be read, is calculated. */
emilmont 1:fdd22bb7aa52 91 readIndex = (int32_t) (S->stateIndex - blockSize) - *pTapDelay++;
emilmont 1:fdd22bb7aa52 92
emilmont 1:fdd22bb7aa52 93 /* Wraparound of readIndex */
emilmont 1:fdd22bb7aa52 94 if(readIndex < 0)
emilmont 1:fdd22bb7aa52 95 {
emilmont 1:fdd22bb7aa52 96 readIndex += (int32_t) delaySize;
emilmont 1:fdd22bb7aa52 97 }
emilmont 1:fdd22bb7aa52 98
emilmont 1:fdd22bb7aa52 99 /* Working pointer for state buffer is updated */
emilmont 1:fdd22bb7aa52 100 py = pState;
emilmont 1:fdd22bb7aa52 101
emilmont 1:fdd22bb7aa52 102 /* blockSize samples are read from the state buffer */
emilmont 1:fdd22bb7aa52 103 arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
emilmont 1:fdd22bb7aa52 104 (int32_t *) pb, (int32_t *) pb, blockSize, 1,
emilmont 1:fdd22bb7aa52 105 blockSize);
emilmont 1:fdd22bb7aa52 106
emilmont 1:fdd22bb7aa52 107 /* Working pointer for the scratch buffer of state values */
emilmont 1:fdd22bb7aa52 108 px = pb;
emilmont 1:fdd22bb7aa52 109
emilmont 1:fdd22bb7aa52 110 /* Working pointer for scratch buffer of output values */
emilmont 1:fdd22bb7aa52 111 pOut = pDst;
emilmont 1:fdd22bb7aa52 112
emilmont 1:fdd22bb7aa52 113
emilmont 1:fdd22bb7aa52 114 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 115
emilmont 1:fdd22bb7aa52 116 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 117
emilmont 1:fdd22bb7aa52 118 /* Loop over the blockSize. Unroll by a factor of 4.
emilmont 1:fdd22bb7aa52 119 * Compute 4 Multiplications at a time. */
emilmont 1:fdd22bb7aa52 120 blkCnt = blockSize >> 2;
emilmont 1:fdd22bb7aa52 121
emilmont 1:fdd22bb7aa52 122 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 123 {
emilmont 1:fdd22bb7aa52 124 /* Perform Multiplications and store in the destination buffer */
emilmont 1:fdd22bb7aa52 125 *pOut++ = (q31_t) (((q63_t) * px++ * coeff) >> 32);
emilmont 1:fdd22bb7aa52 126 *pOut++ = (q31_t) (((q63_t) * px++ * coeff) >> 32);
emilmont 1:fdd22bb7aa52 127 *pOut++ = (q31_t) (((q63_t) * px++ * coeff) >> 32);
emilmont 1:fdd22bb7aa52 128 *pOut++ = (q31_t) (((q63_t) * px++ * coeff) >> 32);
emilmont 1:fdd22bb7aa52 129
emilmont 1:fdd22bb7aa52 130 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 131 blkCnt--;
emilmont 1:fdd22bb7aa52 132 }
emilmont 1:fdd22bb7aa52 133
emilmont 1:fdd22bb7aa52 134 /* If the blockSize is not a multiple of 4,
emilmont 1:fdd22bb7aa52 135 * compute the remaining samples */
emilmont 1:fdd22bb7aa52 136 blkCnt = blockSize % 0x4u;
emilmont 1:fdd22bb7aa52 137
emilmont 1:fdd22bb7aa52 138 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 139 {
emilmont 1:fdd22bb7aa52 140 /* Perform Multiplications and store in the destination buffer */
emilmont 1:fdd22bb7aa52 141 *pOut++ = (q31_t) (((q63_t) * px++ * coeff) >> 32);
emilmont 1:fdd22bb7aa52 142
emilmont 1:fdd22bb7aa52 143 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 144 blkCnt--;
emilmont 1:fdd22bb7aa52 145 }
emilmont 1:fdd22bb7aa52 146
emilmont 1:fdd22bb7aa52 147 /* Load the coefficient value and
emilmont 1:fdd22bb7aa52 148 * increment the coefficient buffer for the next set of state values */
emilmont 1:fdd22bb7aa52 149 coeff = *pCoeffs++;
emilmont 1:fdd22bb7aa52 150
emilmont 1:fdd22bb7aa52 151 /* Read Index, from where the state buffer should be read, is calculated. */
emilmont 1:fdd22bb7aa52 152 readIndex = (int32_t) (S->stateIndex - blockSize) - *pTapDelay++;
emilmont 1:fdd22bb7aa52 153
emilmont 1:fdd22bb7aa52 154 /* Wraparound of readIndex */
emilmont 1:fdd22bb7aa52 155 if(readIndex < 0)
emilmont 1:fdd22bb7aa52 156 {
emilmont 1:fdd22bb7aa52 157 readIndex += (int32_t) delaySize;
emilmont 1:fdd22bb7aa52 158 }
emilmont 1:fdd22bb7aa52 159
emilmont 1:fdd22bb7aa52 160 /* Loop over the number of taps. */
emilmont 1:fdd22bb7aa52 161 tapCnt = (uint32_t) numTaps - 1u;
emilmont 1:fdd22bb7aa52 162
emilmont 1:fdd22bb7aa52 163 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 164 {
emilmont 1:fdd22bb7aa52 165 /* Working pointer for state buffer is updated */
emilmont 1:fdd22bb7aa52 166 py = pState;
emilmont 1:fdd22bb7aa52 167
emilmont 1:fdd22bb7aa52 168 /* blockSize samples are read from the state buffer */
emilmont 1:fdd22bb7aa52 169 arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
emilmont 1:fdd22bb7aa52 170 (int32_t *) pb, (int32_t *) pb, blockSize, 1,
emilmont 1:fdd22bb7aa52 171 blockSize);
emilmont 1:fdd22bb7aa52 172
emilmont 1:fdd22bb7aa52 173 /* Working pointer for the scratch buffer of state values */
emilmont 1:fdd22bb7aa52 174 px = pb;
emilmont 1:fdd22bb7aa52 175
emilmont 1:fdd22bb7aa52 176 /* Working pointer for scratch buffer of output values */
emilmont 1:fdd22bb7aa52 177 pOut = pDst;
emilmont 1:fdd22bb7aa52 178
emilmont 1:fdd22bb7aa52 179 /* Loop over the blockSize. Unroll by a factor of 4.
emilmont 1:fdd22bb7aa52 180 * Compute 4 MACS at a time. */
emilmont 1:fdd22bb7aa52 181 blkCnt = blockSize >> 2;
emilmont 1:fdd22bb7aa52 182
emilmont 1:fdd22bb7aa52 183 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 184 {
emilmont 1:fdd22bb7aa52 185 out = *pOut;
emilmont 1:fdd22bb7aa52 186 out += ((q63_t) * px++ * coeff) >> 32;
emilmont 1:fdd22bb7aa52 187 *pOut++ = (q31_t) (out);
emilmont 1:fdd22bb7aa52 188
emilmont 1:fdd22bb7aa52 189 out = *pOut;
emilmont 1:fdd22bb7aa52 190 out += ((q63_t) * px++ * coeff) >> 32;
emilmont 1:fdd22bb7aa52 191 *pOut++ = (q31_t) (out);
emilmont 1:fdd22bb7aa52 192
emilmont 1:fdd22bb7aa52 193 out = *pOut;
emilmont 1:fdd22bb7aa52 194 out += ((q63_t) * px++ * coeff) >> 32;
emilmont 1:fdd22bb7aa52 195 *pOut++ = (q31_t) (out);
emilmont 1:fdd22bb7aa52 196
emilmont 1:fdd22bb7aa52 197 out = *pOut;
emilmont 1:fdd22bb7aa52 198 out += ((q63_t) * px++ * coeff) >> 32;
emilmont 1:fdd22bb7aa52 199 *pOut++ = (q31_t) (out);
emilmont 1:fdd22bb7aa52 200
emilmont 1:fdd22bb7aa52 201 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 202 blkCnt--;
emilmont 1:fdd22bb7aa52 203 }
emilmont 1:fdd22bb7aa52 204
emilmont 1:fdd22bb7aa52 205 /* If the blockSize is not a multiple of 4,
emilmont 1:fdd22bb7aa52 206 * compute the remaining samples */
emilmont 1:fdd22bb7aa52 207 blkCnt = blockSize % 0x4u;
emilmont 1:fdd22bb7aa52 208
emilmont 1:fdd22bb7aa52 209 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 210 {
emilmont 1:fdd22bb7aa52 211 /* Perform Multiply-Accumulate */
emilmont 1:fdd22bb7aa52 212 out = *pOut;
emilmont 1:fdd22bb7aa52 213 out += ((q63_t) * px++ * coeff) >> 32;
emilmont 1:fdd22bb7aa52 214 *pOut++ = (q31_t) (out);
emilmont 1:fdd22bb7aa52 215
emilmont 1:fdd22bb7aa52 216 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 217 blkCnt--;
emilmont 1:fdd22bb7aa52 218 }
emilmont 1:fdd22bb7aa52 219
emilmont 1:fdd22bb7aa52 220 /* Load the coefficient value and
emilmont 1:fdd22bb7aa52 221 * increment the coefficient buffer for the next set of state values */
emilmont 1:fdd22bb7aa52 222 coeff = *pCoeffs++;
emilmont 1:fdd22bb7aa52 223
emilmont 1:fdd22bb7aa52 224 /* Read Index, from where the state buffer should be read, is calculated. */
emilmont 1:fdd22bb7aa52 225 readIndex = (int32_t) (S->stateIndex - blockSize) - *pTapDelay++;
emilmont 1:fdd22bb7aa52 226
emilmont 1:fdd22bb7aa52 227 /* Wraparound of readIndex */
emilmont 1:fdd22bb7aa52 228 if(readIndex < 0)
emilmont 1:fdd22bb7aa52 229 {
emilmont 1:fdd22bb7aa52 230 readIndex += (int32_t) delaySize;
emilmont 1:fdd22bb7aa52 231 }
emilmont 1:fdd22bb7aa52 232
emilmont 1:fdd22bb7aa52 233 /* Decrement the tap loop counter */
emilmont 1:fdd22bb7aa52 234 tapCnt--;
emilmont 1:fdd22bb7aa52 235 }
emilmont 1:fdd22bb7aa52 236
emilmont 1:fdd22bb7aa52 237 /* Working output pointer is updated */
emilmont 1:fdd22bb7aa52 238 pOut = pDst;
emilmont 1:fdd22bb7aa52 239
emilmont 1:fdd22bb7aa52 240 /* Output is converted into 1.31 format. */
emilmont 1:fdd22bb7aa52 241 /* Loop over the blockSize. Unroll by a factor of 4.
emilmont 1:fdd22bb7aa52 242 * process 4 output samples at a time. */
emilmont 1:fdd22bb7aa52 243 blkCnt = blockSize >> 2;
emilmont 1:fdd22bb7aa52 244
emilmont 1:fdd22bb7aa52 245 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 246 {
emilmont 1:fdd22bb7aa52 247 in = *pOut << 1;
emilmont 1:fdd22bb7aa52 248 *pOut++ = in;
emilmont 1:fdd22bb7aa52 249 in = *pOut << 1;
emilmont 1:fdd22bb7aa52 250 *pOut++ = in;
emilmont 1:fdd22bb7aa52 251 in = *pOut << 1;
emilmont 1:fdd22bb7aa52 252 *pOut++ = in;
emilmont 1:fdd22bb7aa52 253 in = *pOut << 1;
emilmont 1:fdd22bb7aa52 254 *pOut++ = in;
emilmont 1:fdd22bb7aa52 255
emilmont 1:fdd22bb7aa52 256 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 257 blkCnt--;
emilmont 1:fdd22bb7aa52 258 }
emilmont 1:fdd22bb7aa52 259
emilmont 1:fdd22bb7aa52 260 /* If the blockSize is not a multiple of 4,
emilmont 1:fdd22bb7aa52 261 * process the remaining output samples */
emilmont 1:fdd22bb7aa52 262 blkCnt = blockSize % 0x4u;
emilmont 1:fdd22bb7aa52 263
emilmont 1:fdd22bb7aa52 264 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 265 {
emilmont 1:fdd22bb7aa52 266 in = *pOut << 1;
emilmont 1:fdd22bb7aa52 267 *pOut++ = in;
emilmont 1:fdd22bb7aa52 268
emilmont 1:fdd22bb7aa52 269 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 270 blkCnt--;
emilmont 1:fdd22bb7aa52 271 }
emilmont 1:fdd22bb7aa52 272
emilmont 1:fdd22bb7aa52 273 #else
emilmont 1:fdd22bb7aa52 274
emilmont 1:fdd22bb7aa52 275 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 276 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 277
emilmont 1:fdd22bb7aa52 278 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 279 {
emilmont 1:fdd22bb7aa52 280 /* Perform Multiplications and store in the destination buffer */
emilmont 1:fdd22bb7aa52 281 *pOut++ = (q31_t) (((q63_t) * px++ * coeff) >> 32);
emilmont 1:fdd22bb7aa52 282
emilmont 1:fdd22bb7aa52 283 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 284 blkCnt--;
emilmont 1:fdd22bb7aa52 285 }
emilmont 1:fdd22bb7aa52 286
emilmont 1:fdd22bb7aa52 287 /* Load the coefficient value and
emilmont 1:fdd22bb7aa52 288 * increment the coefficient buffer for the next set of state values */
emilmont 1:fdd22bb7aa52 289 coeff = *pCoeffs++;
emilmont 1:fdd22bb7aa52 290
emilmont 1:fdd22bb7aa52 291 /* Read Index, from where the state buffer should be read, is calculated. */
emilmont 1:fdd22bb7aa52 292 readIndex = (int32_t) (S->stateIndex - blockSize) - *pTapDelay++;
emilmont 1:fdd22bb7aa52 293
emilmont 1:fdd22bb7aa52 294 /* Wraparound of readIndex */
emilmont 1:fdd22bb7aa52 295 if(readIndex < 0)
emilmont 1:fdd22bb7aa52 296 {
emilmont 1:fdd22bb7aa52 297 readIndex += (int32_t) delaySize;
emilmont 1:fdd22bb7aa52 298 }
emilmont 1:fdd22bb7aa52 299
emilmont 1:fdd22bb7aa52 300 /* Loop over the number of taps. */
emilmont 1:fdd22bb7aa52 301 tapCnt = (uint32_t) numTaps - 1u;
emilmont 1:fdd22bb7aa52 302
emilmont 1:fdd22bb7aa52 303 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 304 {
emilmont 1:fdd22bb7aa52 305 /* Working pointer for state buffer is updated */
emilmont 1:fdd22bb7aa52 306 py = pState;
emilmont 1:fdd22bb7aa52 307
emilmont 1:fdd22bb7aa52 308 /* blockSize samples are read from the state buffer */
emilmont 1:fdd22bb7aa52 309 arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
emilmont 1:fdd22bb7aa52 310 (int32_t *) pb, (int32_t *) pb, blockSize, 1,
emilmont 1:fdd22bb7aa52 311 blockSize);
emilmont 1:fdd22bb7aa52 312
emilmont 1:fdd22bb7aa52 313 /* Working pointer for the scratch buffer of state values */
emilmont 1:fdd22bb7aa52 314 px = pb;
emilmont 1:fdd22bb7aa52 315
emilmont 1:fdd22bb7aa52 316 /* Working pointer for scratch buffer of output values */
emilmont 1:fdd22bb7aa52 317 pOut = pDst;
emilmont 1:fdd22bb7aa52 318
emilmont 1:fdd22bb7aa52 319 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 320
emilmont 1:fdd22bb7aa52 321 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 322 {
emilmont 1:fdd22bb7aa52 323 /* Perform Multiply-Accumulate */
emilmont 1:fdd22bb7aa52 324 out = *pOut;
emilmont 1:fdd22bb7aa52 325 out += ((q63_t) * px++ * coeff) >> 32;
emilmont 1:fdd22bb7aa52 326 *pOut++ = (q31_t) (out);
emilmont 1:fdd22bb7aa52 327
emilmont 1:fdd22bb7aa52 328 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 329 blkCnt--;
emilmont 1:fdd22bb7aa52 330 }
emilmont 1:fdd22bb7aa52 331
emilmont 1:fdd22bb7aa52 332 /* Load the coefficient value and
emilmont 1:fdd22bb7aa52 333 * increment the coefficient buffer for the next set of state values */
emilmont 1:fdd22bb7aa52 334 coeff = *pCoeffs++;
emilmont 1:fdd22bb7aa52 335
emilmont 1:fdd22bb7aa52 336 /* Read Index, from where the state buffer should be read, is calculated. */
emilmont 1:fdd22bb7aa52 337 readIndex = (int32_t) (S->stateIndex - blockSize) - *pTapDelay++;
emilmont 1:fdd22bb7aa52 338
emilmont 1:fdd22bb7aa52 339 /* Wraparound of readIndex */
emilmont 1:fdd22bb7aa52 340 if(readIndex < 0)
emilmont 1:fdd22bb7aa52 341 {
emilmont 1:fdd22bb7aa52 342 readIndex += (int32_t) delaySize;
emilmont 1:fdd22bb7aa52 343 }
emilmont 1:fdd22bb7aa52 344
emilmont 1:fdd22bb7aa52 345 /* Decrement the tap loop counter */
emilmont 1:fdd22bb7aa52 346 tapCnt--;
emilmont 1:fdd22bb7aa52 347 }
emilmont 1:fdd22bb7aa52 348
emilmont 1:fdd22bb7aa52 349 /* Working output pointer is updated */
emilmont 1:fdd22bb7aa52 350 pOut = pDst;
emilmont 1:fdd22bb7aa52 351
emilmont 1:fdd22bb7aa52 352 /* Output is converted into 1.31 format. */
emilmont 1:fdd22bb7aa52 353 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 354
emilmont 1:fdd22bb7aa52 355 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 356 {
emilmont 1:fdd22bb7aa52 357 in = *pOut << 1;
emilmont 1:fdd22bb7aa52 358 *pOut++ = in;
emilmont 1:fdd22bb7aa52 359
emilmont 1:fdd22bb7aa52 360 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 361 blkCnt--;
emilmont 1:fdd22bb7aa52 362 }
emilmont 1:fdd22bb7aa52 363
emilmont 1:fdd22bb7aa52 364 #endif /* #ifndef ARM_MATH_CM0 */
emilmont 1:fdd22bb7aa52 365
emilmont 1:fdd22bb7aa52 366 }
emilmont 1:fdd22bb7aa52 367
emilmont 1:fdd22bb7aa52 368 /**
emilmont 1:fdd22bb7aa52 369 * @} end of FIR_Sparse group
emilmont 1:fdd22bb7aa52 370 */