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_q7.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?

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 2:da51fb522205 5 * $Revision: V1.1.0
emilmont 1:fdd22bb7aa52 6 *
emilmont 2:da51fb522205 7 * Project: CMSIS DSP Library
emilmont 1:fdd22bb7aa52 8 * Title: arm_fir_q7.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 1:fdd22bb7aa52 10 * Description: Q7 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 0.0.5 2010/04/26
emilmont 2:da51fb522205 33 * incorporated review comments and updated with latest CMSIS layer
emilmont 1:fdd22bb7aa52 34 *
emilmont 1:fdd22bb7aa52 35 * Version 0.0.3 2010/03/10
emilmont 1:fdd22bb7aa52 36 * Initial version
emilmont 1:fdd22bb7aa52 37 * -------------------------------------------------------------------- */
emilmont 1:fdd22bb7aa52 38
emilmont 1:fdd22bb7aa52 39 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 40
emilmont 1:fdd22bb7aa52 41 /**
emilmont 1:fdd22bb7aa52 42 * @ingroup groupFilters
emilmont 1:fdd22bb7aa52 43 */
emilmont 1:fdd22bb7aa52 44
emilmont 1:fdd22bb7aa52 45 /**
emilmont 1:fdd22bb7aa52 46 * @addtogroup FIR
emilmont 1:fdd22bb7aa52 47 * @{
emilmont 1:fdd22bb7aa52 48 */
emilmont 1:fdd22bb7aa52 49
emilmont 1:fdd22bb7aa52 50 /**
emilmont 1:fdd22bb7aa52 51 * @param[in] *S points to an instance of the Q7 FIR filter structure.
emilmont 1:fdd22bb7aa52 52 * @param[in] *pSrc points to the block of input data.
emilmont 1:fdd22bb7aa52 53 * @param[out] *pDst points to the block of output data.
emilmont 1:fdd22bb7aa52 54 * @param[in] blockSize number of samples to process per call.
emilmont 2:da51fb522205 55 * @return none.
emilmont 1:fdd22bb7aa52 56 *
emilmont 1:fdd22bb7aa52 57 * <b>Scaling and Overflow Behavior:</b>
emilmont 1:fdd22bb7aa52 58 * \par
emilmont 1:fdd22bb7aa52 59 * The function is implemented using a 32-bit internal accumulator.
emilmont 1:fdd22bb7aa52 60 * Both coefficients and state variables are represented in 1.7 format and multiplications yield a 2.14 result.
emilmont 1:fdd22bb7aa52 61 * The 2.14 intermediate results are accumulated in a 32-bit accumulator in 18.14 format.
emilmont 1:fdd22bb7aa52 62 * There is no risk of internal overflow with this approach and the full precision of intermediate multiplications is preserved.
emilmont 1:fdd22bb7aa52 63 * The accumulator is converted to 18.7 format by discarding the low 7 bits.
emilmont 1:fdd22bb7aa52 64 * Finally, the result is truncated to 1.7 format.
emilmont 1:fdd22bb7aa52 65 */
emilmont 1:fdd22bb7aa52 66
emilmont 1:fdd22bb7aa52 67 void arm_fir_q7(
emilmont 1:fdd22bb7aa52 68 const arm_fir_instance_q7 * S,
emilmont 1:fdd22bb7aa52 69 q7_t * pSrc,
emilmont 1:fdd22bb7aa52 70 q7_t * pDst,
emilmont 1:fdd22bb7aa52 71 uint32_t blockSize)
emilmont 1:fdd22bb7aa52 72 {
emilmont 1:fdd22bb7aa52 73
emilmont 1:fdd22bb7aa52 74 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 75
emilmont 1:fdd22bb7aa52 76 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 77
emilmont 1:fdd22bb7aa52 78 q7_t *pState = S->pState; /* State pointer */
emilmont 1:fdd22bb7aa52 79 q7_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emilmont 1:fdd22bb7aa52 80 q7_t *pStateCurnt; /* Points to the current sample of the state */
emilmont 1:fdd22bb7aa52 81 q7_t x0, x1, x2, x3; /* Temporary variables to hold state */
emilmont 1:fdd22bb7aa52 82 q7_t c0; /* Temporary variable to hold coefficient value */
emilmont 1:fdd22bb7aa52 83 q7_t *px; /* Temporary pointer for state */
emilmont 1:fdd22bb7aa52 84 q7_t *pb; /* Temporary pointer for coefficient buffer */
emilmont 1:fdd22bb7aa52 85 q31_t acc0, acc1, acc2, acc3; /* Accumulators */
emilmont 1:fdd22bb7aa52 86 uint32_t numTaps = S->numTaps; /* Number of filter coefficients in the filter */
emilmont 1:fdd22bb7aa52 87 uint32_t i, tapCnt, blkCnt; /* Loop counters */
emilmont 1:fdd22bb7aa52 88
emilmont 1:fdd22bb7aa52 89 /* S->pState points to state array which contains previous frame (numTaps - 1) samples */
emilmont 1:fdd22bb7aa52 90 /* pStateCurnt points to the location where the new input data should be written */
emilmont 1:fdd22bb7aa52 91 pStateCurnt = &(S->pState[(numTaps - 1u)]);
emilmont 1:fdd22bb7aa52 92
emilmont 1:fdd22bb7aa52 93 /* Apply loop unrolling and compute 4 output values simultaneously.
emilmont 1:fdd22bb7aa52 94 * The variables acc0 ... acc3 hold output values that are being computed:
emilmont 1:fdd22bb7aa52 95 *
emilmont 1:fdd22bb7aa52 96 * acc0 = b[numTaps-1] * x[n-numTaps-1] + b[numTaps-2] * x[n-numTaps-2] + b[numTaps-3] * x[n-numTaps-3] +...+ b[0] * x[0]
emilmont 1:fdd22bb7aa52 97 * acc1 = b[numTaps-1] * x[n-numTaps] + b[numTaps-2] * x[n-numTaps-1] + b[numTaps-3] * x[n-numTaps-2] +...+ b[0] * x[1]
emilmont 1:fdd22bb7aa52 98 * acc2 = b[numTaps-1] * x[n-numTaps+1] + b[numTaps-2] * x[n-numTaps] + b[numTaps-3] * x[n-numTaps-1] +...+ b[0] * x[2]
emilmont 1:fdd22bb7aa52 99 * acc3 = b[numTaps-1] * x[n-numTaps+2] + b[numTaps-2] * x[n-numTaps+1] + b[numTaps-3] * x[n-numTaps] +...+ b[0] * x[3]
emilmont 1:fdd22bb7aa52 100 */
emilmont 1:fdd22bb7aa52 101 blkCnt = blockSize >> 2;
emilmont 1:fdd22bb7aa52 102
emilmont 1:fdd22bb7aa52 103 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
emilmont 1:fdd22bb7aa52 104 ** a second loop below computes the remaining 1 to 3 samples. */
emilmont 1:fdd22bb7aa52 105 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 106 {
emilmont 1:fdd22bb7aa52 107 /* Copy four new input samples into the state buffer */
emilmont 1:fdd22bb7aa52 108 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 109 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 110 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 111 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 112
emilmont 1:fdd22bb7aa52 113 /* Set all accumulators to zero */
emilmont 1:fdd22bb7aa52 114 acc0 = 0;
emilmont 1:fdd22bb7aa52 115 acc1 = 0;
emilmont 1:fdd22bb7aa52 116 acc2 = 0;
emilmont 1:fdd22bb7aa52 117 acc3 = 0;
emilmont 1:fdd22bb7aa52 118
emilmont 1:fdd22bb7aa52 119 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 120 px = pState;
emilmont 1:fdd22bb7aa52 121
emilmont 1:fdd22bb7aa52 122 /* Initialize coefficient pointer */
emilmont 1:fdd22bb7aa52 123 pb = pCoeffs;
emilmont 1:fdd22bb7aa52 124
emilmont 1:fdd22bb7aa52 125 /* Read the first three samples from the state buffer:
emilmont 1:fdd22bb7aa52 126 * x[n-numTaps], x[n-numTaps-1], x[n-numTaps-2] */
emilmont 1:fdd22bb7aa52 127 x0 = *(px++);
emilmont 1:fdd22bb7aa52 128 x1 = *(px++);
emilmont 1:fdd22bb7aa52 129 x2 = *(px++);
emilmont 1:fdd22bb7aa52 130
emilmont 1:fdd22bb7aa52 131 /* Loop unrolling. Process 4 taps at a time. */
emilmont 1:fdd22bb7aa52 132 tapCnt = numTaps >> 2;
emilmont 1:fdd22bb7aa52 133 i = tapCnt;
emilmont 1:fdd22bb7aa52 134
emilmont 1:fdd22bb7aa52 135 while(i > 0u)
emilmont 1:fdd22bb7aa52 136 {
emilmont 1:fdd22bb7aa52 137 /* Read the b[numTaps] coefficient */
emilmont 1:fdd22bb7aa52 138 c0 = *(pb++);
emilmont 1:fdd22bb7aa52 139
emilmont 1:fdd22bb7aa52 140 /* Read x[n-numTaps-3] sample */
emilmont 1:fdd22bb7aa52 141 x3 = *(px++);
emilmont 1:fdd22bb7aa52 142
emilmont 1:fdd22bb7aa52 143 /* acc0 += b[numTaps] * x[n-numTaps] */
emilmont 1:fdd22bb7aa52 144 acc0 += ((q15_t) x0 * c0);
emilmont 1:fdd22bb7aa52 145
emilmont 1:fdd22bb7aa52 146 /* acc1 += b[numTaps] * x[n-numTaps-1] */
emilmont 1:fdd22bb7aa52 147 acc1 += ((q15_t) x1 * c0);
emilmont 1:fdd22bb7aa52 148
emilmont 1:fdd22bb7aa52 149 /* acc2 += b[numTaps] * x[n-numTaps-2] */
emilmont 1:fdd22bb7aa52 150 acc2 += ((q15_t) x2 * c0);
emilmont 1:fdd22bb7aa52 151
emilmont 1:fdd22bb7aa52 152 /* acc3 += b[numTaps] * x[n-numTaps-3] */
emilmont 1:fdd22bb7aa52 153 acc3 += ((q15_t) x3 * c0);
emilmont 1:fdd22bb7aa52 154
emilmont 1:fdd22bb7aa52 155 /* Read the b[numTaps-1] coefficient */
emilmont 1:fdd22bb7aa52 156 c0 = *(pb++);
emilmont 1:fdd22bb7aa52 157
emilmont 1:fdd22bb7aa52 158 /* Read x[n-numTaps-4] sample */
emilmont 1:fdd22bb7aa52 159 x0 = *(px++);
emilmont 1:fdd22bb7aa52 160
emilmont 1:fdd22bb7aa52 161 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 162 acc0 += ((q15_t) x1 * c0);
emilmont 1:fdd22bb7aa52 163 acc1 += ((q15_t) x2 * c0);
emilmont 1:fdd22bb7aa52 164 acc2 += ((q15_t) x3 * c0);
emilmont 1:fdd22bb7aa52 165 acc3 += ((q15_t) x0 * c0);
emilmont 1:fdd22bb7aa52 166
emilmont 1:fdd22bb7aa52 167 /* Read the b[numTaps-2] coefficient */
emilmont 1:fdd22bb7aa52 168 c0 = *(pb++);
emilmont 1:fdd22bb7aa52 169
emilmont 1:fdd22bb7aa52 170 /* Read x[n-numTaps-5] sample */
emilmont 1:fdd22bb7aa52 171 x1 = *(px++);
emilmont 1:fdd22bb7aa52 172
emilmont 1:fdd22bb7aa52 173 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 174 acc0 += ((q15_t) x2 * c0);
emilmont 1:fdd22bb7aa52 175 acc1 += ((q15_t) x3 * c0);
emilmont 1:fdd22bb7aa52 176 acc2 += ((q15_t) x0 * c0);
emilmont 1:fdd22bb7aa52 177 acc3 += ((q15_t) x1 * c0);
emilmont 1:fdd22bb7aa52 178 /* Read the b[numTaps-3] coefficients */
emilmont 1:fdd22bb7aa52 179 c0 = *(pb++);
emilmont 1:fdd22bb7aa52 180
emilmont 1:fdd22bb7aa52 181 /* Read x[n-numTaps-6] sample */
emilmont 1:fdd22bb7aa52 182 x2 = *(px++);
emilmont 1:fdd22bb7aa52 183
emilmont 1:fdd22bb7aa52 184 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 185 acc0 += ((q15_t) x3 * c0);
emilmont 1:fdd22bb7aa52 186 acc1 += ((q15_t) x0 * c0);
emilmont 1:fdd22bb7aa52 187 acc2 += ((q15_t) x1 * c0);
emilmont 1:fdd22bb7aa52 188 acc3 += ((q15_t) x2 * c0);
emilmont 1:fdd22bb7aa52 189 i--;
emilmont 1:fdd22bb7aa52 190 }
emilmont 1:fdd22bb7aa52 191
emilmont 1:fdd22bb7aa52 192 /* If the filter length is not a multiple of 4, compute the remaining filter taps */
emilmont 1:fdd22bb7aa52 193
emilmont 1:fdd22bb7aa52 194 i = numTaps - (tapCnt * 4u);
emilmont 1:fdd22bb7aa52 195 while(i > 0u)
emilmont 1:fdd22bb7aa52 196 {
emilmont 1:fdd22bb7aa52 197 /* Read coefficients */
emilmont 1:fdd22bb7aa52 198 c0 = *(pb++);
emilmont 1:fdd22bb7aa52 199
emilmont 1:fdd22bb7aa52 200 /* Fetch 1 state variable */
emilmont 1:fdd22bb7aa52 201 x3 = *(px++);
emilmont 1:fdd22bb7aa52 202
emilmont 1:fdd22bb7aa52 203 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 204 acc0 += ((q15_t) x0 * c0);
emilmont 1:fdd22bb7aa52 205 acc1 += ((q15_t) x1 * c0);
emilmont 1:fdd22bb7aa52 206 acc2 += ((q15_t) x2 * c0);
emilmont 1:fdd22bb7aa52 207 acc3 += ((q15_t) x3 * c0);
emilmont 1:fdd22bb7aa52 208
emilmont 1:fdd22bb7aa52 209 /* Reuse the present sample states for next sample */
emilmont 1:fdd22bb7aa52 210 x0 = x1;
emilmont 1:fdd22bb7aa52 211 x1 = x2;
emilmont 1:fdd22bb7aa52 212 x2 = x3;
emilmont 1:fdd22bb7aa52 213
emilmont 1:fdd22bb7aa52 214 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 215 i--;
emilmont 1:fdd22bb7aa52 216 }
emilmont 1:fdd22bb7aa52 217
emilmont 1:fdd22bb7aa52 218 /* Advance the state pointer by 4 to process the next group of 4 samples */
emilmont 1:fdd22bb7aa52 219 pState = pState + 4;
emilmont 1:fdd22bb7aa52 220
emilmont 1:fdd22bb7aa52 221 /* The results in the 4 accumulators are in 2.62 format. Convert to 1.31
emilmont 1:fdd22bb7aa52 222 ** Then store the 4 outputs in the destination buffer. */
emilmont 1:fdd22bb7aa52 223 acc0 = __SSAT((acc0 >> 7u), 8);
emilmont 1:fdd22bb7aa52 224 *pDst++ = acc0;
emilmont 1:fdd22bb7aa52 225 acc1 = __SSAT((acc1 >> 7u), 8);
emilmont 1:fdd22bb7aa52 226 *pDst++ = acc1;
emilmont 1:fdd22bb7aa52 227 acc2 = __SSAT((acc2 >> 7u), 8);
emilmont 1:fdd22bb7aa52 228 *pDst++ = acc2;
emilmont 1:fdd22bb7aa52 229 acc3 = __SSAT((acc3 >> 7u), 8);
emilmont 1:fdd22bb7aa52 230 *pDst++ = acc3;
emilmont 1:fdd22bb7aa52 231
emilmont 1:fdd22bb7aa52 232 /* Decrement the samples loop counter */
emilmont 1:fdd22bb7aa52 233 blkCnt--;
emilmont 1:fdd22bb7aa52 234 }
emilmont 1:fdd22bb7aa52 235
emilmont 1:fdd22bb7aa52 236
emilmont 1:fdd22bb7aa52 237 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
emilmont 1:fdd22bb7aa52 238 ** No loop unrolling is used. */
emilmont 1:fdd22bb7aa52 239 blkCnt = blockSize % 4u;
emilmont 1:fdd22bb7aa52 240
emilmont 1:fdd22bb7aa52 241 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 242 {
emilmont 1:fdd22bb7aa52 243 /* Copy one sample at a time into state buffer */
emilmont 1:fdd22bb7aa52 244 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 245
emilmont 1:fdd22bb7aa52 246 /* Set the accumulator to zero */
emilmont 1:fdd22bb7aa52 247 acc0 = 0;
emilmont 1:fdd22bb7aa52 248
emilmont 1:fdd22bb7aa52 249 /* Initialize state pointer */
emilmont 1:fdd22bb7aa52 250 px = pState;
emilmont 1:fdd22bb7aa52 251
emilmont 1:fdd22bb7aa52 252 /* Initialize Coefficient pointer */
emilmont 1:fdd22bb7aa52 253 pb = (pCoeffs);
emilmont 1:fdd22bb7aa52 254
emilmont 1:fdd22bb7aa52 255 i = numTaps;
emilmont 1:fdd22bb7aa52 256
emilmont 1:fdd22bb7aa52 257 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 258 do
emilmont 1:fdd22bb7aa52 259 {
emilmont 1:fdd22bb7aa52 260 acc0 += (q15_t) * (px++) * (*(pb++));
emilmont 1:fdd22bb7aa52 261 i--;
emilmont 1:fdd22bb7aa52 262 } while(i > 0u);
emilmont 1:fdd22bb7aa52 263
emilmont 1:fdd22bb7aa52 264 /* The result is in 2.14 format. Convert to 1.7
emilmont 1:fdd22bb7aa52 265 ** Then store the output in the destination buffer. */
emilmont 1:fdd22bb7aa52 266 *pDst++ = __SSAT((acc0 >> 7u), 8);
emilmont 1:fdd22bb7aa52 267
emilmont 1:fdd22bb7aa52 268 /* Advance state pointer by 1 for the next sample */
emilmont 1:fdd22bb7aa52 269 pState = pState + 1;
emilmont 1:fdd22bb7aa52 270
emilmont 1:fdd22bb7aa52 271 /* Decrement the samples loop counter */
emilmont 1:fdd22bb7aa52 272 blkCnt--;
emilmont 1:fdd22bb7aa52 273 }
emilmont 1:fdd22bb7aa52 274
emilmont 1:fdd22bb7aa52 275 /* Processing is complete.
emilmont 1:fdd22bb7aa52 276 ** Now copy the last numTaps - 1 samples to the satrt of the state buffer.
emilmont 1:fdd22bb7aa52 277 ** This prepares the state buffer for the next function call. */
emilmont 1:fdd22bb7aa52 278
emilmont 1:fdd22bb7aa52 279 /* Points to the start of the state buffer */
emilmont 1:fdd22bb7aa52 280 pStateCurnt = S->pState;
emilmont 1:fdd22bb7aa52 281
emilmont 1:fdd22bb7aa52 282 tapCnt = (numTaps - 1u) >> 2u;
emilmont 1:fdd22bb7aa52 283
emilmont 1:fdd22bb7aa52 284 /* copy data */
emilmont 1:fdd22bb7aa52 285 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 286 {
emilmont 1:fdd22bb7aa52 287 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 288 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 289 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 290 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 291
emilmont 1:fdd22bb7aa52 292 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 293 tapCnt--;
emilmont 1:fdd22bb7aa52 294 }
emilmont 1:fdd22bb7aa52 295
emilmont 1:fdd22bb7aa52 296 /* Calculate remaining number of copies */
emilmont 1:fdd22bb7aa52 297 tapCnt = (numTaps - 1u) % 0x4u;
emilmont 1:fdd22bb7aa52 298
emilmont 1:fdd22bb7aa52 299 /* Copy the remaining q31_t data */
emilmont 1:fdd22bb7aa52 300 while(tapCnt > 0u)
emilmont 1:fdd22bb7aa52 301 {
emilmont 1:fdd22bb7aa52 302 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 303
emilmont 1:fdd22bb7aa52 304 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 305 tapCnt--;
emilmont 1:fdd22bb7aa52 306 }
emilmont 1:fdd22bb7aa52 307
emilmont 1:fdd22bb7aa52 308 #else
emilmont 1:fdd22bb7aa52 309
emilmont 1:fdd22bb7aa52 310 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 311
emilmont 1:fdd22bb7aa52 312 uint32_t numTaps = S->numTaps; /* Number of taps in the filter */
emilmont 1:fdd22bb7aa52 313 uint32_t i, blkCnt; /* Loop counters */
emilmont 1:fdd22bb7aa52 314 q7_t *pState = S->pState; /* State pointer */
emilmont 1:fdd22bb7aa52 315 q7_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */
emilmont 1:fdd22bb7aa52 316 q7_t *px, *pb; /* Temporary pointers to state and coeff */
emilmont 1:fdd22bb7aa52 317 q31_t acc = 0; /* Accumlator */
emilmont 1:fdd22bb7aa52 318 q7_t *pStateCurnt; /* Points to the current sample of the state */
emilmont 1:fdd22bb7aa52 319
emilmont 1:fdd22bb7aa52 320
emilmont 1:fdd22bb7aa52 321 /* S->pState points to state array which contains previous frame (numTaps - 1) samples */
emilmont 1:fdd22bb7aa52 322 /* pStateCurnt points to the location where the new input data should be written */
emilmont 1:fdd22bb7aa52 323 pStateCurnt = S->pState + (numTaps - 1u);
emilmont 1:fdd22bb7aa52 324
emilmont 1:fdd22bb7aa52 325 /* Initialize blkCnt with blockSize */
emilmont 1:fdd22bb7aa52 326 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 327
emilmont 1:fdd22bb7aa52 328 /* Perform filtering upto BlockSize - BlockSize%4 */
emilmont 1:fdd22bb7aa52 329 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 330 {
emilmont 1:fdd22bb7aa52 331 /* Copy one sample at a time into state buffer */
emilmont 1:fdd22bb7aa52 332 *pStateCurnt++ = *pSrc++;
emilmont 1:fdd22bb7aa52 333
emilmont 1:fdd22bb7aa52 334 /* Set accumulator to zero */
emilmont 1:fdd22bb7aa52 335 acc = 0;
emilmont 1:fdd22bb7aa52 336
emilmont 1:fdd22bb7aa52 337 /* Initialize state pointer of type q7 */
emilmont 1:fdd22bb7aa52 338 px = pState;
emilmont 1:fdd22bb7aa52 339
emilmont 1:fdd22bb7aa52 340 /* Initialize coeff pointer of type q7 */
emilmont 1:fdd22bb7aa52 341 pb = pCoeffs;
emilmont 1:fdd22bb7aa52 342
emilmont 1:fdd22bb7aa52 343
emilmont 1:fdd22bb7aa52 344 i = numTaps;
emilmont 1:fdd22bb7aa52 345
emilmont 1:fdd22bb7aa52 346 while(i > 0u)
emilmont 1:fdd22bb7aa52 347 {
emilmont 1:fdd22bb7aa52 348 /* acc = b[numTaps-1] * x[n-numTaps-1] + b[numTaps-2] * x[n-numTaps-2] + b[numTaps-3] * x[n-numTaps-3] +...+ b[0] * x[0] */
emilmont 1:fdd22bb7aa52 349 acc += (q15_t) * px++ * *pb++;
emilmont 1:fdd22bb7aa52 350 i--;
emilmont 1:fdd22bb7aa52 351 }
emilmont 1:fdd22bb7aa52 352
emilmont 1:fdd22bb7aa52 353 /* Store the 1.7 format filter output in destination buffer */
emilmont 1:fdd22bb7aa52 354 *pDst++ = (q7_t) __SSAT((acc >> 7), 8);
emilmont 1:fdd22bb7aa52 355
emilmont 1:fdd22bb7aa52 356 /* Advance the state pointer by 1 to process the next sample */
emilmont 1:fdd22bb7aa52 357 pState = pState + 1;
emilmont 1:fdd22bb7aa52 358
emilmont 1:fdd22bb7aa52 359 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 360 blkCnt--;
emilmont 1:fdd22bb7aa52 361 }
emilmont 1:fdd22bb7aa52 362
emilmont 1:fdd22bb7aa52 363 /* Processing is complete.
emilmont 1:fdd22bb7aa52 364 ** Now copy the last numTaps - 1 samples to the satrt of the state buffer.
emilmont 1:fdd22bb7aa52 365 ** This prepares the state buffer for the next function call. */
emilmont 1:fdd22bb7aa52 366
emilmont 1:fdd22bb7aa52 367
emilmont 1:fdd22bb7aa52 368 /* Points to the start of the state buffer */
emilmont 1:fdd22bb7aa52 369 pStateCurnt = S->pState;
emilmont 1:fdd22bb7aa52 370
emilmont 1:fdd22bb7aa52 371
emilmont 1:fdd22bb7aa52 372 /* Copy numTaps number of values */
emilmont 1:fdd22bb7aa52 373 i = (numTaps - 1u);
emilmont 1:fdd22bb7aa52 374
emilmont 1:fdd22bb7aa52 375 /* Copy q7_t data */
emilmont 1:fdd22bb7aa52 376 while(i > 0u)
emilmont 1:fdd22bb7aa52 377 {
emilmont 1:fdd22bb7aa52 378 *pStateCurnt++ = *pState++;
emilmont 1:fdd22bb7aa52 379 i--;
emilmont 1:fdd22bb7aa52 380 }
emilmont 1:fdd22bb7aa52 381
emilmont 1:fdd22bb7aa52 382 #endif /* #ifndef ARM_MATH_CM0 */
emilmont 1:fdd22bb7aa52 383
emilmont 1:fdd22bb7aa52 384 }
emilmont 1:fdd22bb7aa52 385
emilmont 1:fdd22bb7aa52 386 /**
emilmont 1:fdd22bb7aa52 387 * @} end of FIR group
emilmont 1:fdd22bb7aa52 388 */