V4.0.1 of the ARM CMSIS DSP libraries. Note that arm_bitreversal2.s, arm_cfft_f32.c and arm_rfft_fast_f32.c had to be removed. arm_bitreversal2.s will not assemble with the online tools. So, the fast f32 FFT functions are not yet available. All the other FFT functions are available.
Dependents: MPU9150_Example fir_f32 fir_f32 MPU9150_nucleo_noni2cdev ... more
FilteringFunctions/arm_fir_decimate_q15.c@0:3d9c67d97d6f, 2014-07-28 (annotated)
- Committer:
- emh203
- Date:
- Mon Jul 28 15:03:15 2014 +0000
- Revision:
- 0:3d9c67d97d6f
1st working commit. Had to remove arm_bitreversal2.s arm_cfft_f32.c and arm_rfft_fast_f32.c. The .s will not assemble. For now I removed these functions so we could at least have a library for the other functions.
Who changed what in which revision?
| User | Revision | Line number | New contents of line |
|---|---|---|---|
| emh203 | 0:3d9c67d97d6f | 1 | /* ---------------------------------------------------------------------- |
| emh203 | 0:3d9c67d97d6f | 2 | * Copyright (C) 2010-2014 ARM Limited. All rights reserved. |
| emh203 | 0:3d9c67d97d6f | 3 | * |
| emh203 | 0:3d9c67d97d6f | 4 | * $Date: 12. March 2014 |
| emh203 | 0:3d9c67d97d6f | 5 | * $Revision: V1.4.3 |
| emh203 | 0:3d9c67d97d6f | 6 | * |
| emh203 | 0:3d9c67d97d6f | 7 | * Project: CMSIS DSP Library |
| emh203 | 0:3d9c67d97d6f | 8 | * Title: arm_fir_decimate_q15.c |
| emh203 | 0:3d9c67d97d6f | 9 | * |
| emh203 | 0:3d9c67d97d6f | 10 | * Description: Q15 FIR Decimator. |
| emh203 | 0:3d9c67d97d6f | 11 | * |
| emh203 | 0:3d9c67d97d6f | 12 | * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0 |
| emh203 | 0:3d9c67d97d6f | 13 | * |
| emh203 | 0:3d9c67d97d6f | 14 | * Redistribution and use in source and binary forms, with or without |
| emh203 | 0:3d9c67d97d6f | 15 | * modification, are permitted provided that the following conditions |
| emh203 | 0:3d9c67d97d6f | 16 | * are met: |
| emh203 | 0:3d9c67d97d6f | 17 | * - Redistributions of source code must retain the above copyright |
| emh203 | 0:3d9c67d97d6f | 18 | * notice, this list of conditions and the following disclaimer. |
| emh203 | 0:3d9c67d97d6f | 19 | * - Redistributions in binary form must reproduce the above copyright |
| emh203 | 0:3d9c67d97d6f | 20 | * notice, this list of conditions and the following disclaimer in |
| emh203 | 0:3d9c67d97d6f | 21 | * the documentation and/or other materials provided with the |
| emh203 | 0:3d9c67d97d6f | 22 | * distribution. |
| emh203 | 0:3d9c67d97d6f | 23 | * - Neither the name of ARM LIMITED nor the names of its contributors |
| emh203 | 0:3d9c67d97d6f | 24 | * may be used to endorse or promote products derived from this |
| emh203 | 0:3d9c67d97d6f | 25 | * software without specific prior written permission. |
| emh203 | 0:3d9c67d97d6f | 26 | * |
| emh203 | 0:3d9c67d97d6f | 27 | * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
| emh203 | 0:3d9c67d97d6f | 28 | * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
| emh203 | 0:3d9c67d97d6f | 29 | * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS |
| emh203 | 0:3d9c67d97d6f | 30 | * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE |
| emh203 | 0:3d9c67d97d6f | 31 | * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, |
| emh203 | 0:3d9c67d97d6f | 32 | * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, |
| emh203 | 0:3d9c67d97d6f | 33 | * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; |
| emh203 | 0:3d9c67d97d6f | 34 | * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER |
| emh203 | 0:3d9c67d97d6f | 35 | * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
| emh203 | 0:3d9c67d97d6f | 36 | * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN |
| emh203 | 0:3d9c67d97d6f | 37 | * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE |
| emh203 | 0:3d9c67d97d6f | 38 | * POSSIBILITY OF SUCH DAMAGE. |
| emh203 | 0:3d9c67d97d6f | 39 | * -------------------------------------------------------------------- */ |
| emh203 | 0:3d9c67d97d6f | 40 | |
| emh203 | 0:3d9c67d97d6f | 41 | #include "arm_math.h" |
| emh203 | 0:3d9c67d97d6f | 42 | |
| emh203 | 0:3d9c67d97d6f | 43 | /** |
| emh203 | 0:3d9c67d97d6f | 44 | * @ingroup groupFilters |
| emh203 | 0:3d9c67d97d6f | 45 | */ |
| emh203 | 0:3d9c67d97d6f | 46 | |
| emh203 | 0:3d9c67d97d6f | 47 | /** |
| emh203 | 0:3d9c67d97d6f | 48 | * @addtogroup FIR_decimate |
| emh203 | 0:3d9c67d97d6f | 49 | * @{ |
| emh203 | 0:3d9c67d97d6f | 50 | */ |
| emh203 | 0:3d9c67d97d6f | 51 | |
| emh203 | 0:3d9c67d97d6f | 52 | /** |
| emh203 | 0:3d9c67d97d6f | 53 | * @brief Processing function for the Q15 FIR decimator. |
| emh203 | 0:3d9c67d97d6f | 54 | * @param[in] *S points to an instance of the Q15 FIR decimator structure. |
| emh203 | 0:3d9c67d97d6f | 55 | * @param[in] *pSrc points to the block of input data. |
| emh203 | 0:3d9c67d97d6f | 56 | * @param[out] *pDst points to the location where the output result is written. |
| emh203 | 0:3d9c67d97d6f | 57 | * @param[in] blockSize number of input samples to process per call. |
| emh203 | 0:3d9c67d97d6f | 58 | * @return none. |
| emh203 | 0:3d9c67d97d6f | 59 | * |
| emh203 | 0:3d9c67d97d6f | 60 | * <b>Scaling and Overflow Behavior:</b> |
| emh203 | 0:3d9c67d97d6f | 61 | * \par |
| emh203 | 0:3d9c67d97d6f | 62 | * The function is implemented using a 64-bit internal accumulator. |
| emh203 | 0:3d9c67d97d6f | 63 | * Both coefficients and state variables are represented in 1.15 format and multiplications yield a 2.30 result. |
| emh203 | 0:3d9c67d97d6f | 64 | * The 2.30 intermediate results are accumulated in a 64-bit accumulator in 34.30 format. |
| emh203 | 0:3d9c67d97d6f | 65 | * There is no risk of internal overflow with this approach and the full precision of intermediate multiplications is preserved. |
| emh203 | 0:3d9c67d97d6f | 66 | * After all additions have been performed, the accumulator is truncated to 34.15 format by discarding low 15 bits. |
| emh203 | 0:3d9c67d97d6f | 67 | * Lastly, the accumulator is saturated to yield a result in 1.15 format. |
| emh203 | 0:3d9c67d97d6f | 68 | * |
| emh203 | 0:3d9c67d97d6f | 69 | * \par |
| emh203 | 0:3d9c67d97d6f | 70 | * Refer to the function <code>arm_fir_decimate_fast_q15()</code> for a faster but less precise implementation of this function for Cortex-M3 and Cortex-M4. |
| emh203 | 0:3d9c67d97d6f | 71 | */ |
| emh203 | 0:3d9c67d97d6f | 72 | |
| emh203 | 0:3d9c67d97d6f | 73 | #ifndef ARM_MATH_CM0_FAMILY |
| emh203 | 0:3d9c67d97d6f | 74 | |
| emh203 | 0:3d9c67d97d6f | 75 | #ifndef UNALIGNED_SUPPORT_DISABLE |
| emh203 | 0:3d9c67d97d6f | 76 | |
| emh203 | 0:3d9c67d97d6f | 77 | void arm_fir_decimate_q15( |
| emh203 | 0:3d9c67d97d6f | 78 | const arm_fir_decimate_instance_q15 * S, |
| emh203 | 0:3d9c67d97d6f | 79 | q15_t * pSrc, |
| emh203 | 0:3d9c67d97d6f | 80 | q15_t * pDst, |
| emh203 | 0:3d9c67d97d6f | 81 | uint32_t blockSize) |
| emh203 | 0:3d9c67d97d6f | 82 | { |
| emh203 | 0:3d9c67d97d6f | 83 | q15_t *pState = S->pState; /* State pointer */ |
| emh203 | 0:3d9c67d97d6f | 84 | q15_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */ |
| emh203 | 0:3d9c67d97d6f | 85 | q15_t *pStateCurnt; /* Points to the current sample of the state */ |
| emh203 | 0:3d9c67d97d6f | 86 | q15_t *px; /* Temporary pointer for state buffer */ |
| emh203 | 0:3d9c67d97d6f | 87 | q15_t *pb; /* Temporary pointer coefficient buffer */ |
| emh203 | 0:3d9c67d97d6f | 88 | q31_t x0, x1, c0, c1; /* Temporary variables to hold state and coefficient values */ |
| emh203 | 0:3d9c67d97d6f | 89 | q63_t sum0; /* Accumulators */ |
| emh203 | 0:3d9c67d97d6f | 90 | q63_t acc0, acc1; |
| emh203 | 0:3d9c67d97d6f | 91 | q15_t *px0, *px1; |
| emh203 | 0:3d9c67d97d6f | 92 | uint32_t blkCntN3; |
| emh203 | 0:3d9c67d97d6f | 93 | uint32_t numTaps = S->numTaps; /* Number of taps */ |
| emh203 | 0:3d9c67d97d6f | 94 | uint32_t i, blkCnt, tapCnt, outBlockSize = blockSize / S->M; /* Loop counters */ |
| emh203 | 0:3d9c67d97d6f | 95 | |
| emh203 | 0:3d9c67d97d6f | 96 | |
| emh203 | 0:3d9c67d97d6f | 97 | /* S->pState buffer contains previous frame (numTaps - 1) samples */ |
| emh203 | 0:3d9c67d97d6f | 98 | /* pStateCurnt points to the location where the new input data should be written */ |
| emh203 | 0:3d9c67d97d6f | 99 | pStateCurnt = S->pState + (numTaps - 1u); |
| emh203 | 0:3d9c67d97d6f | 100 | |
| emh203 | 0:3d9c67d97d6f | 101 | |
| emh203 | 0:3d9c67d97d6f | 102 | /* Total number of output samples to be computed */ |
| emh203 | 0:3d9c67d97d6f | 103 | blkCnt = outBlockSize / 2; |
| emh203 | 0:3d9c67d97d6f | 104 | blkCntN3 = outBlockSize - (2 * blkCnt); |
| emh203 | 0:3d9c67d97d6f | 105 | |
| emh203 | 0:3d9c67d97d6f | 106 | |
| emh203 | 0:3d9c67d97d6f | 107 | while(blkCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 108 | { |
| emh203 | 0:3d9c67d97d6f | 109 | /* Copy decimation factor number of new input samples into the state buffer */ |
| emh203 | 0:3d9c67d97d6f | 110 | i = 2 * S->M; |
| emh203 | 0:3d9c67d97d6f | 111 | |
| emh203 | 0:3d9c67d97d6f | 112 | do |
| emh203 | 0:3d9c67d97d6f | 113 | { |
| emh203 | 0:3d9c67d97d6f | 114 | *pStateCurnt++ = *pSrc++; |
| emh203 | 0:3d9c67d97d6f | 115 | |
| emh203 | 0:3d9c67d97d6f | 116 | } while(--i); |
| emh203 | 0:3d9c67d97d6f | 117 | |
| emh203 | 0:3d9c67d97d6f | 118 | /* Set accumulator to zero */ |
| emh203 | 0:3d9c67d97d6f | 119 | acc0 = 0; |
| emh203 | 0:3d9c67d97d6f | 120 | acc1 = 0; |
| emh203 | 0:3d9c67d97d6f | 121 | |
| emh203 | 0:3d9c67d97d6f | 122 | /* Initialize state pointer */ |
| emh203 | 0:3d9c67d97d6f | 123 | px0 = pState; |
| emh203 | 0:3d9c67d97d6f | 124 | |
| emh203 | 0:3d9c67d97d6f | 125 | px1 = pState + S->M; |
| emh203 | 0:3d9c67d97d6f | 126 | |
| emh203 | 0:3d9c67d97d6f | 127 | |
| emh203 | 0:3d9c67d97d6f | 128 | /* Initialize coeff pointer */ |
| emh203 | 0:3d9c67d97d6f | 129 | pb = pCoeffs; |
| emh203 | 0:3d9c67d97d6f | 130 | |
| emh203 | 0:3d9c67d97d6f | 131 | /* Loop unrolling. Process 4 taps at a time. */ |
| emh203 | 0:3d9c67d97d6f | 132 | tapCnt = numTaps >> 2; |
| emh203 | 0:3d9c67d97d6f | 133 | |
| emh203 | 0:3d9c67d97d6f | 134 | /* Loop over the number of taps. Unroll by a factor of 4. |
| emh203 | 0:3d9c67d97d6f | 135 | ** Repeat until we've computed numTaps-4 coefficients. */ |
| emh203 | 0:3d9c67d97d6f | 136 | while(tapCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 137 | { |
| emh203 | 0:3d9c67d97d6f | 138 | /* Read the Read b[numTaps-1] and b[numTaps-2] coefficients */ |
| emh203 | 0:3d9c67d97d6f | 139 | c0 = *__SIMD32(pb)++; |
| emh203 | 0:3d9c67d97d6f | 140 | |
| emh203 | 0:3d9c67d97d6f | 141 | /* Read x[n-numTaps-1] and x[n-numTaps-2]sample */ |
| emh203 | 0:3d9c67d97d6f | 142 | x0 = *__SIMD32(px0)++; |
| emh203 | 0:3d9c67d97d6f | 143 | |
| emh203 | 0:3d9c67d97d6f | 144 | x1 = *__SIMD32(px1)++; |
| emh203 | 0:3d9c67d97d6f | 145 | |
| emh203 | 0:3d9c67d97d6f | 146 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 147 | acc0 = __SMLALD(x0, c0, acc0); |
| emh203 | 0:3d9c67d97d6f | 148 | |
| emh203 | 0:3d9c67d97d6f | 149 | acc1 = __SMLALD(x1, c0, acc1); |
| emh203 | 0:3d9c67d97d6f | 150 | |
| emh203 | 0:3d9c67d97d6f | 151 | /* Read the b[numTaps-3] and b[numTaps-4] coefficient */ |
| emh203 | 0:3d9c67d97d6f | 152 | c0 = *__SIMD32(pb)++; |
| emh203 | 0:3d9c67d97d6f | 153 | |
| emh203 | 0:3d9c67d97d6f | 154 | /* Read x[n-numTaps-2] and x[n-numTaps-3] sample */ |
| emh203 | 0:3d9c67d97d6f | 155 | x0 = *__SIMD32(px0)++; |
| emh203 | 0:3d9c67d97d6f | 156 | |
| emh203 | 0:3d9c67d97d6f | 157 | x1 = *__SIMD32(px1)++; |
| emh203 | 0:3d9c67d97d6f | 158 | |
| emh203 | 0:3d9c67d97d6f | 159 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 160 | acc0 = __SMLALD(x0, c0, acc0); |
| emh203 | 0:3d9c67d97d6f | 161 | |
| emh203 | 0:3d9c67d97d6f | 162 | acc1 = __SMLALD(x1, c0, acc1); |
| emh203 | 0:3d9c67d97d6f | 163 | |
| emh203 | 0:3d9c67d97d6f | 164 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 165 | tapCnt--; |
| emh203 | 0:3d9c67d97d6f | 166 | } |
| emh203 | 0:3d9c67d97d6f | 167 | |
| emh203 | 0:3d9c67d97d6f | 168 | /* If the filter length is not a multiple of 4, compute the remaining filter taps */ |
| emh203 | 0:3d9c67d97d6f | 169 | tapCnt = numTaps % 0x4u; |
| emh203 | 0:3d9c67d97d6f | 170 | |
| emh203 | 0:3d9c67d97d6f | 171 | while(tapCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 172 | { |
| emh203 | 0:3d9c67d97d6f | 173 | /* Read coefficients */ |
| emh203 | 0:3d9c67d97d6f | 174 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 175 | |
| emh203 | 0:3d9c67d97d6f | 176 | /* Fetch 1 state variable */ |
| emh203 | 0:3d9c67d97d6f | 177 | x0 = *px0++; |
| emh203 | 0:3d9c67d97d6f | 178 | |
| emh203 | 0:3d9c67d97d6f | 179 | x1 = *px1++; |
| emh203 | 0:3d9c67d97d6f | 180 | |
| emh203 | 0:3d9c67d97d6f | 181 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 182 | acc0 = __SMLALD(x0, c0, acc0); |
| emh203 | 0:3d9c67d97d6f | 183 | acc1 = __SMLALD(x1, c0, acc1); |
| emh203 | 0:3d9c67d97d6f | 184 | |
| emh203 | 0:3d9c67d97d6f | 185 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 186 | tapCnt--; |
| emh203 | 0:3d9c67d97d6f | 187 | } |
| emh203 | 0:3d9c67d97d6f | 188 | |
| emh203 | 0:3d9c67d97d6f | 189 | /* Advance the state pointer by the decimation factor |
| emh203 | 0:3d9c67d97d6f | 190 | * to process the next group of decimation factor number samples */ |
| emh203 | 0:3d9c67d97d6f | 191 | pState = pState + S->M * 2; |
| emh203 | 0:3d9c67d97d6f | 192 | |
| emh203 | 0:3d9c67d97d6f | 193 | /* Store filter output, smlad returns the values in 2.14 format */ |
| emh203 | 0:3d9c67d97d6f | 194 | /* so downsacle by 15 to get output in 1.15 */ |
| emh203 | 0:3d9c67d97d6f | 195 | *pDst++ = (q15_t) (__SSAT((acc0 >> 15), 16)); |
| emh203 | 0:3d9c67d97d6f | 196 | *pDst++ = (q15_t) (__SSAT((acc1 >> 15), 16)); |
| emh203 | 0:3d9c67d97d6f | 197 | |
| emh203 | 0:3d9c67d97d6f | 198 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 199 | blkCnt--; |
| emh203 | 0:3d9c67d97d6f | 200 | } |
| emh203 | 0:3d9c67d97d6f | 201 | |
| emh203 | 0:3d9c67d97d6f | 202 | |
| emh203 | 0:3d9c67d97d6f | 203 | |
| emh203 | 0:3d9c67d97d6f | 204 | while(blkCntN3 > 0u) |
| emh203 | 0:3d9c67d97d6f | 205 | { |
| emh203 | 0:3d9c67d97d6f | 206 | /* Copy decimation factor number of new input samples into the state buffer */ |
| emh203 | 0:3d9c67d97d6f | 207 | i = S->M; |
| emh203 | 0:3d9c67d97d6f | 208 | |
| emh203 | 0:3d9c67d97d6f | 209 | do |
| emh203 | 0:3d9c67d97d6f | 210 | { |
| emh203 | 0:3d9c67d97d6f | 211 | *pStateCurnt++ = *pSrc++; |
| emh203 | 0:3d9c67d97d6f | 212 | |
| emh203 | 0:3d9c67d97d6f | 213 | } while(--i); |
| emh203 | 0:3d9c67d97d6f | 214 | |
| emh203 | 0:3d9c67d97d6f | 215 | /*Set sum to zero */ |
| emh203 | 0:3d9c67d97d6f | 216 | sum0 = 0; |
| emh203 | 0:3d9c67d97d6f | 217 | |
| emh203 | 0:3d9c67d97d6f | 218 | /* Initialize state pointer */ |
| emh203 | 0:3d9c67d97d6f | 219 | px = pState; |
| emh203 | 0:3d9c67d97d6f | 220 | |
| emh203 | 0:3d9c67d97d6f | 221 | /* Initialize coeff pointer */ |
| emh203 | 0:3d9c67d97d6f | 222 | pb = pCoeffs; |
| emh203 | 0:3d9c67d97d6f | 223 | |
| emh203 | 0:3d9c67d97d6f | 224 | /* Loop unrolling. Process 4 taps at a time. */ |
| emh203 | 0:3d9c67d97d6f | 225 | tapCnt = numTaps >> 2; |
| emh203 | 0:3d9c67d97d6f | 226 | |
| emh203 | 0:3d9c67d97d6f | 227 | /* Loop over the number of taps. Unroll by a factor of 4. |
| emh203 | 0:3d9c67d97d6f | 228 | ** Repeat until we've computed numTaps-4 coefficients. */ |
| emh203 | 0:3d9c67d97d6f | 229 | while(tapCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 230 | { |
| emh203 | 0:3d9c67d97d6f | 231 | /* Read the Read b[numTaps-1] and b[numTaps-2] coefficients */ |
| emh203 | 0:3d9c67d97d6f | 232 | c0 = *__SIMD32(pb)++; |
| emh203 | 0:3d9c67d97d6f | 233 | |
| emh203 | 0:3d9c67d97d6f | 234 | /* Read x[n-numTaps-1] and x[n-numTaps-2]sample */ |
| emh203 | 0:3d9c67d97d6f | 235 | x0 = *__SIMD32(px)++; |
| emh203 | 0:3d9c67d97d6f | 236 | |
| emh203 | 0:3d9c67d97d6f | 237 | /* Read the b[numTaps-3] and b[numTaps-4] coefficient */ |
| emh203 | 0:3d9c67d97d6f | 238 | c1 = *__SIMD32(pb)++; |
| emh203 | 0:3d9c67d97d6f | 239 | |
| emh203 | 0:3d9c67d97d6f | 240 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 241 | sum0 = __SMLALD(x0, c0, sum0); |
| emh203 | 0:3d9c67d97d6f | 242 | |
| emh203 | 0:3d9c67d97d6f | 243 | /* Read x[n-numTaps-2] and x[n-numTaps-3] sample */ |
| emh203 | 0:3d9c67d97d6f | 244 | x0 = *__SIMD32(px)++; |
| emh203 | 0:3d9c67d97d6f | 245 | |
| emh203 | 0:3d9c67d97d6f | 246 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 247 | sum0 = __SMLALD(x0, c1, sum0); |
| emh203 | 0:3d9c67d97d6f | 248 | |
| emh203 | 0:3d9c67d97d6f | 249 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 250 | tapCnt--; |
| emh203 | 0:3d9c67d97d6f | 251 | } |
| emh203 | 0:3d9c67d97d6f | 252 | |
| emh203 | 0:3d9c67d97d6f | 253 | /* If the filter length is not a multiple of 4, compute the remaining filter taps */ |
| emh203 | 0:3d9c67d97d6f | 254 | tapCnt = numTaps % 0x4u; |
| emh203 | 0:3d9c67d97d6f | 255 | |
| emh203 | 0:3d9c67d97d6f | 256 | while(tapCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 257 | { |
| emh203 | 0:3d9c67d97d6f | 258 | /* Read coefficients */ |
| emh203 | 0:3d9c67d97d6f | 259 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 260 | |
| emh203 | 0:3d9c67d97d6f | 261 | /* Fetch 1 state variable */ |
| emh203 | 0:3d9c67d97d6f | 262 | x0 = *px++; |
| emh203 | 0:3d9c67d97d6f | 263 | |
| emh203 | 0:3d9c67d97d6f | 264 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 265 | sum0 = __SMLALD(x0, c0, sum0); |
| emh203 | 0:3d9c67d97d6f | 266 | |
| emh203 | 0:3d9c67d97d6f | 267 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 268 | tapCnt--; |
| emh203 | 0:3d9c67d97d6f | 269 | } |
| emh203 | 0:3d9c67d97d6f | 270 | |
| emh203 | 0:3d9c67d97d6f | 271 | /* Advance the state pointer by the decimation factor |
| emh203 | 0:3d9c67d97d6f | 272 | * to process the next group of decimation factor number samples */ |
| emh203 | 0:3d9c67d97d6f | 273 | pState = pState + S->M; |
| emh203 | 0:3d9c67d97d6f | 274 | |
| emh203 | 0:3d9c67d97d6f | 275 | /* Store filter output, smlad returns the values in 2.14 format */ |
| emh203 | 0:3d9c67d97d6f | 276 | /* so downsacle by 15 to get output in 1.15 */ |
| emh203 | 0:3d9c67d97d6f | 277 | *pDst++ = (q15_t) (__SSAT((sum0 >> 15), 16)); |
| emh203 | 0:3d9c67d97d6f | 278 | |
| emh203 | 0:3d9c67d97d6f | 279 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 280 | blkCntN3--; |
| emh203 | 0:3d9c67d97d6f | 281 | } |
| emh203 | 0:3d9c67d97d6f | 282 | |
| emh203 | 0:3d9c67d97d6f | 283 | /* Processing is complete. |
| emh203 | 0:3d9c67d97d6f | 284 | ** Now copy the last numTaps - 1 samples to the satrt of the state buffer. |
| emh203 | 0:3d9c67d97d6f | 285 | ** This prepares the state buffer for the next function call. */ |
| emh203 | 0:3d9c67d97d6f | 286 | |
| emh203 | 0:3d9c67d97d6f | 287 | /* Points to the start of the state buffer */ |
| emh203 | 0:3d9c67d97d6f | 288 | pStateCurnt = S->pState; |
| emh203 | 0:3d9c67d97d6f | 289 | |
| emh203 | 0:3d9c67d97d6f | 290 | i = (numTaps - 1u) >> 2u; |
| emh203 | 0:3d9c67d97d6f | 291 | |
| emh203 | 0:3d9c67d97d6f | 292 | /* copy data */ |
| emh203 | 0:3d9c67d97d6f | 293 | while(i > 0u) |
| emh203 | 0:3d9c67d97d6f | 294 | { |
| emh203 | 0:3d9c67d97d6f | 295 | *__SIMD32(pStateCurnt)++ = *__SIMD32(pState)++; |
| emh203 | 0:3d9c67d97d6f | 296 | *__SIMD32(pStateCurnt)++ = *__SIMD32(pState)++; |
| emh203 | 0:3d9c67d97d6f | 297 | |
| emh203 | 0:3d9c67d97d6f | 298 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 299 | i--; |
| emh203 | 0:3d9c67d97d6f | 300 | } |
| emh203 | 0:3d9c67d97d6f | 301 | |
| emh203 | 0:3d9c67d97d6f | 302 | i = (numTaps - 1u) % 0x04u; |
| emh203 | 0:3d9c67d97d6f | 303 | |
| emh203 | 0:3d9c67d97d6f | 304 | /* copy data */ |
| emh203 | 0:3d9c67d97d6f | 305 | while(i > 0u) |
| emh203 | 0:3d9c67d97d6f | 306 | { |
| emh203 | 0:3d9c67d97d6f | 307 | *pStateCurnt++ = *pState++; |
| emh203 | 0:3d9c67d97d6f | 308 | |
| emh203 | 0:3d9c67d97d6f | 309 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 310 | i--; |
| emh203 | 0:3d9c67d97d6f | 311 | } |
| emh203 | 0:3d9c67d97d6f | 312 | } |
| emh203 | 0:3d9c67d97d6f | 313 | |
| emh203 | 0:3d9c67d97d6f | 314 | #else |
| emh203 | 0:3d9c67d97d6f | 315 | |
| emh203 | 0:3d9c67d97d6f | 316 | |
| emh203 | 0:3d9c67d97d6f | 317 | void arm_fir_decimate_q15( |
| emh203 | 0:3d9c67d97d6f | 318 | const arm_fir_decimate_instance_q15 * S, |
| emh203 | 0:3d9c67d97d6f | 319 | q15_t * pSrc, |
| emh203 | 0:3d9c67d97d6f | 320 | q15_t * pDst, |
| emh203 | 0:3d9c67d97d6f | 321 | uint32_t blockSize) |
| emh203 | 0:3d9c67d97d6f | 322 | { |
| emh203 | 0:3d9c67d97d6f | 323 | q15_t *pState = S->pState; /* State pointer */ |
| emh203 | 0:3d9c67d97d6f | 324 | q15_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */ |
| emh203 | 0:3d9c67d97d6f | 325 | q15_t *pStateCurnt; /* Points to the current sample of the state */ |
| emh203 | 0:3d9c67d97d6f | 326 | q15_t *px; /* Temporary pointer for state buffer */ |
| emh203 | 0:3d9c67d97d6f | 327 | q15_t *pb; /* Temporary pointer coefficient buffer */ |
| emh203 | 0:3d9c67d97d6f | 328 | q15_t x0, x1, c0; /* Temporary variables to hold state and coefficient values */ |
| emh203 | 0:3d9c67d97d6f | 329 | q63_t sum0; /* Accumulators */ |
| emh203 | 0:3d9c67d97d6f | 330 | q63_t acc0, acc1; |
| emh203 | 0:3d9c67d97d6f | 331 | q15_t *px0, *px1; |
| emh203 | 0:3d9c67d97d6f | 332 | uint32_t blkCntN3; |
| emh203 | 0:3d9c67d97d6f | 333 | uint32_t numTaps = S->numTaps; /* Number of taps */ |
| emh203 | 0:3d9c67d97d6f | 334 | uint32_t i, blkCnt, tapCnt, outBlockSize = blockSize / S->M; /* Loop counters */ |
| emh203 | 0:3d9c67d97d6f | 335 | |
| emh203 | 0:3d9c67d97d6f | 336 | |
| emh203 | 0:3d9c67d97d6f | 337 | /* S->pState buffer contains previous frame (numTaps - 1) samples */ |
| emh203 | 0:3d9c67d97d6f | 338 | /* pStateCurnt points to the location where the new input data should be written */ |
| emh203 | 0:3d9c67d97d6f | 339 | pStateCurnt = S->pState + (numTaps - 1u); |
| emh203 | 0:3d9c67d97d6f | 340 | |
| emh203 | 0:3d9c67d97d6f | 341 | |
| emh203 | 0:3d9c67d97d6f | 342 | /* Total number of output samples to be computed */ |
| emh203 | 0:3d9c67d97d6f | 343 | blkCnt = outBlockSize / 2; |
| emh203 | 0:3d9c67d97d6f | 344 | blkCntN3 = outBlockSize - (2 * blkCnt); |
| emh203 | 0:3d9c67d97d6f | 345 | |
| emh203 | 0:3d9c67d97d6f | 346 | while(blkCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 347 | { |
| emh203 | 0:3d9c67d97d6f | 348 | /* Copy decimation factor number of new input samples into the state buffer */ |
| emh203 | 0:3d9c67d97d6f | 349 | i = 2 * S->M; |
| emh203 | 0:3d9c67d97d6f | 350 | |
| emh203 | 0:3d9c67d97d6f | 351 | do |
| emh203 | 0:3d9c67d97d6f | 352 | { |
| emh203 | 0:3d9c67d97d6f | 353 | *pStateCurnt++ = *pSrc++; |
| emh203 | 0:3d9c67d97d6f | 354 | |
| emh203 | 0:3d9c67d97d6f | 355 | } while(--i); |
| emh203 | 0:3d9c67d97d6f | 356 | |
| emh203 | 0:3d9c67d97d6f | 357 | /* Set accumulator to zero */ |
| emh203 | 0:3d9c67d97d6f | 358 | acc0 = 0; |
| emh203 | 0:3d9c67d97d6f | 359 | acc1 = 0; |
| emh203 | 0:3d9c67d97d6f | 360 | |
| emh203 | 0:3d9c67d97d6f | 361 | /* Initialize state pointer */ |
| emh203 | 0:3d9c67d97d6f | 362 | px0 = pState; |
| emh203 | 0:3d9c67d97d6f | 363 | |
| emh203 | 0:3d9c67d97d6f | 364 | px1 = pState + S->M; |
| emh203 | 0:3d9c67d97d6f | 365 | |
| emh203 | 0:3d9c67d97d6f | 366 | |
| emh203 | 0:3d9c67d97d6f | 367 | /* Initialize coeff pointer */ |
| emh203 | 0:3d9c67d97d6f | 368 | pb = pCoeffs; |
| emh203 | 0:3d9c67d97d6f | 369 | |
| emh203 | 0:3d9c67d97d6f | 370 | /* Loop unrolling. Process 4 taps at a time. */ |
| emh203 | 0:3d9c67d97d6f | 371 | tapCnt = numTaps >> 2; |
| emh203 | 0:3d9c67d97d6f | 372 | |
| emh203 | 0:3d9c67d97d6f | 373 | /* Loop over the number of taps. Unroll by a factor of 4. |
| emh203 | 0:3d9c67d97d6f | 374 | ** Repeat until we've computed numTaps-4 coefficients. */ |
| emh203 | 0:3d9c67d97d6f | 375 | while(tapCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 376 | { |
| emh203 | 0:3d9c67d97d6f | 377 | /* Read the Read b[numTaps-1] coefficients */ |
| emh203 | 0:3d9c67d97d6f | 378 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 379 | |
| emh203 | 0:3d9c67d97d6f | 380 | /* Read x[n-numTaps-1] for sample 0 and for sample 1 */ |
| emh203 | 0:3d9c67d97d6f | 381 | x0 = *px0++; |
| emh203 | 0:3d9c67d97d6f | 382 | x1 = *px1++; |
| emh203 | 0:3d9c67d97d6f | 383 | |
| emh203 | 0:3d9c67d97d6f | 384 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 385 | acc0 += x0 * c0; |
| emh203 | 0:3d9c67d97d6f | 386 | acc1 += x1 * c0; |
| emh203 | 0:3d9c67d97d6f | 387 | |
| emh203 | 0:3d9c67d97d6f | 388 | /* Read the b[numTaps-2] coefficient */ |
| emh203 | 0:3d9c67d97d6f | 389 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 390 | |
| emh203 | 0:3d9c67d97d6f | 391 | /* Read x[n-numTaps-2] for sample 0 and sample 1 */ |
| emh203 | 0:3d9c67d97d6f | 392 | x0 = *px0++; |
| emh203 | 0:3d9c67d97d6f | 393 | x1 = *px1++; |
| emh203 | 0:3d9c67d97d6f | 394 | |
| emh203 | 0:3d9c67d97d6f | 395 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 396 | acc0 += x0 * c0; |
| emh203 | 0:3d9c67d97d6f | 397 | acc1 += x1 * c0; |
| emh203 | 0:3d9c67d97d6f | 398 | |
| emh203 | 0:3d9c67d97d6f | 399 | /* Read the b[numTaps-3] coefficients */ |
| emh203 | 0:3d9c67d97d6f | 400 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 401 | |
| emh203 | 0:3d9c67d97d6f | 402 | /* Read x[n-numTaps-3] for sample 0 and sample 1 */ |
| emh203 | 0:3d9c67d97d6f | 403 | x0 = *px0++; |
| emh203 | 0:3d9c67d97d6f | 404 | x1 = *px1++; |
| emh203 | 0:3d9c67d97d6f | 405 | |
| emh203 | 0:3d9c67d97d6f | 406 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 407 | acc0 += x0 * c0; |
| emh203 | 0:3d9c67d97d6f | 408 | acc1 += x1 * c0; |
| emh203 | 0:3d9c67d97d6f | 409 | |
| emh203 | 0:3d9c67d97d6f | 410 | /* Read the b[numTaps-4] coefficient */ |
| emh203 | 0:3d9c67d97d6f | 411 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 412 | |
| emh203 | 0:3d9c67d97d6f | 413 | /* Read x[n-numTaps-4] for sample 0 and sample 1 */ |
| emh203 | 0:3d9c67d97d6f | 414 | x0 = *px0++; |
| emh203 | 0:3d9c67d97d6f | 415 | x1 = *px1++; |
| emh203 | 0:3d9c67d97d6f | 416 | |
| emh203 | 0:3d9c67d97d6f | 417 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 418 | acc0 += x0 * c0; |
| emh203 | 0:3d9c67d97d6f | 419 | acc1 += x1 * c0; |
| emh203 | 0:3d9c67d97d6f | 420 | |
| emh203 | 0:3d9c67d97d6f | 421 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 422 | tapCnt--; |
| emh203 | 0:3d9c67d97d6f | 423 | } |
| emh203 | 0:3d9c67d97d6f | 424 | |
| emh203 | 0:3d9c67d97d6f | 425 | /* If the filter length is not a multiple of 4, compute the remaining filter taps */ |
| emh203 | 0:3d9c67d97d6f | 426 | tapCnt = numTaps % 0x4u; |
| emh203 | 0:3d9c67d97d6f | 427 | |
| emh203 | 0:3d9c67d97d6f | 428 | while(tapCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 429 | { |
| emh203 | 0:3d9c67d97d6f | 430 | /* Read coefficients */ |
| emh203 | 0:3d9c67d97d6f | 431 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 432 | |
| emh203 | 0:3d9c67d97d6f | 433 | /* Fetch 1 state variable */ |
| emh203 | 0:3d9c67d97d6f | 434 | x0 = *px0++; |
| emh203 | 0:3d9c67d97d6f | 435 | x1 = *px1++; |
| emh203 | 0:3d9c67d97d6f | 436 | |
| emh203 | 0:3d9c67d97d6f | 437 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 438 | acc0 += x0 * c0; |
| emh203 | 0:3d9c67d97d6f | 439 | acc1 += x1 * c0; |
| emh203 | 0:3d9c67d97d6f | 440 | |
| emh203 | 0:3d9c67d97d6f | 441 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 442 | tapCnt--; |
| emh203 | 0:3d9c67d97d6f | 443 | } |
| emh203 | 0:3d9c67d97d6f | 444 | |
| emh203 | 0:3d9c67d97d6f | 445 | /* Advance the state pointer by the decimation factor |
| emh203 | 0:3d9c67d97d6f | 446 | * to process the next group of decimation factor number samples */ |
| emh203 | 0:3d9c67d97d6f | 447 | pState = pState + S->M * 2; |
| emh203 | 0:3d9c67d97d6f | 448 | |
| emh203 | 0:3d9c67d97d6f | 449 | /* Store filter output, smlad returns the values in 2.14 format */ |
| emh203 | 0:3d9c67d97d6f | 450 | /* so downsacle by 15 to get output in 1.15 */ |
| emh203 | 0:3d9c67d97d6f | 451 | |
| emh203 | 0:3d9c67d97d6f | 452 | *pDst++ = (q15_t) (__SSAT((acc0 >> 15), 16)); |
| emh203 | 0:3d9c67d97d6f | 453 | *pDst++ = (q15_t) (__SSAT((acc1 >> 15), 16)); |
| emh203 | 0:3d9c67d97d6f | 454 | |
| emh203 | 0:3d9c67d97d6f | 455 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 456 | blkCnt--; |
| emh203 | 0:3d9c67d97d6f | 457 | } |
| emh203 | 0:3d9c67d97d6f | 458 | |
| emh203 | 0:3d9c67d97d6f | 459 | while(blkCntN3 > 0u) |
| emh203 | 0:3d9c67d97d6f | 460 | { |
| emh203 | 0:3d9c67d97d6f | 461 | /* Copy decimation factor number of new input samples into the state buffer */ |
| emh203 | 0:3d9c67d97d6f | 462 | i = S->M; |
| emh203 | 0:3d9c67d97d6f | 463 | |
| emh203 | 0:3d9c67d97d6f | 464 | do |
| emh203 | 0:3d9c67d97d6f | 465 | { |
| emh203 | 0:3d9c67d97d6f | 466 | *pStateCurnt++ = *pSrc++; |
| emh203 | 0:3d9c67d97d6f | 467 | |
| emh203 | 0:3d9c67d97d6f | 468 | } while(--i); |
| emh203 | 0:3d9c67d97d6f | 469 | |
| emh203 | 0:3d9c67d97d6f | 470 | /*Set sum to zero */ |
| emh203 | 0:3d9c67d97d6f | 471 | sum0 = 0; |
| emh203 | 0:3d9c67d97d6f | 472 | |
| emh203 | 0:3d9c67d97d6f | 473 | /* Initialize state pointer */ |
| emh203 | 0:3d9c67d97d6f | 474 | px = pState; |
| emh203 | 0:3d9c67d97d6f | 475 | |
| emh203 | 0:3d9c67d97d6f | 476 | /* Initialize coeff pointer */ |
| emh203 | 0:3d9c67d97d6f | 477 | pb = pCoeffs; |
| emh203 | 0:3d9c67d97d6f | 478 | |
| emh203 | 0:3d9c67d97d6f | 479 | /* Loop unrolling. Process 4 taps at a time. */ |
| emh203 | 0:3d9c67d97d6f | 480 | tapCnt = numTaps >> 2; |
| emh203 | 0:3d9c67d97d6f | 481 | |
| emh203 | 0:3d9c67d97d6f | 482 | /* Loop over the number of taps. Unroll by a factor of 4. |
| emh203 | 0:3d9c67d97d6f | 483 | ** Repeat until we've computed numTaps-4 coefficients. */ |
| emh203 | 0:3d9c67d97d6f | 484 | while(tapCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 485 | { |
| emh203 | 0:3d9c67d97d6f | 486 | /* Read the Read b[numTaps-1] coefficients */ |
| emh203 | 0:3d9c67d97d6f | 487 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 488 | |
| emh203 | 0:3d9c67d97d6f | 489 | /* Read x[n-numTaps-1] and sample */ |
| emh203 | 0:3d9c67d97d6f | 490 | x0 = *px++; |
| emh203 | 0:3d9c67d97d6f | 491 | |
| emh203 | 0:3d9c67d97d6f | 492 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 493 | sum0 += x0 * c0; |
| emh203 | 0:3d9c67d97d6f | 494 | |
| emh203 | 0:3d9c67d97d6f | 495 | /* Read the b[numTaps-2] coefficient */ |
| emh203 | 0:3d9c67d97d6f | 496 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 497 | |
| emh203 | 0:3d9c67d97d6f | 498 | /* Read x[n-numTaps-2] and sample */ |
| emh203 | 0:3d9c67d97d6f | 499 | x0 = *px++; |
| emh203 | 0:3d9c67d97d6f | 500 | |
| emh203 | 0:3d9c67d97d6f | 501 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 502 | sum0 += x0 * c0; |
| emh203 | 0:3d9c67d97d6f | 503 | |
| emh203 | 0:3d9c67d97d6f | 504 | /* Read the b[numTaps-3] coefficients */ |
| emh203 | 0:3d9c67d97d6f | 505 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 506 | |
| emh203 | 0:3d9c67d97d6f | 507 | /* Read x[n-numTaps-3] sample */ |
| emh203 | 0:3d9c67d97d6f | 508 | x0 = *px++; |
| emh203 | 0:3d9c67d97d6f | 509 | |
| emh203 | 0:3d9c67d97d6f | 510 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 511 | sum0 += x0 * c0; |
| emh203 | 0:3d9c67d97d6f | 512 | |
| emh203 | 0:3d9c67d97d6f | 513 | /* Read the b[numTaps-4] coefficient */ |
| emh203 | 0:3d9c67d97d6f | 514 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 515 | |
| emh203 | 0:3d9c67d97d6f | 516 | /* Read x[n-numTaps-4] sample */ |
| emh203 | 0:3d9c67d97d6f | 517 | x0 = *px++; |
| emh203 | 0:3d9c67d97d6f | 518 | |
| emh203 | 0:3d9c67d97d6f | 519 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 520 | sum0 += x0 * c0; |
| emh203 | 0:3d9c67d97d6f | 521 | |
| emh203 | 0:3d9c67d97d6f | 522 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 523 | tapCnt--; |
| emh203 | 0:3d9c67d97d6f | 524 | } |
| emh203 | 0:3d9c67d97d6f | 525 | |
| emh203 | 0:3d9c67d97d6f | 526 | /* If the filter length is not a multiple of 4, compute the remaining filter taps */ |
| emh203 | 0:3d9c67d97d6f | 527 | tapCnt = numTaps % 0x4u; |
| emh203 | 0:3d9c67d97d6f | 528 | |
| emh203 | 0:3d9c67d97d6f | 529 | while(tapCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 530 | { |
| emh203 | 0:3d9c67d97d6f | 531 | /* Read coefficients */ |
| emh203 | 0:3d9c67d97d6f | 532 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 533 | |
| emh203 | 0:3d9c67d97d6f | 534 | /* Fetch 1 state variable */ |
| emh203 | 0:3d9c67d97d6f | 535 | x0 = *px++; |
| emh203 | 0:3d9c67d97d6f | 536 | |
| emh203 | 0:3d9c67d97d6f | 537 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 538 | sum0 += x0 * c0; |
| emh203 | 0:3d9c67d97d6f | 539 | |
| emh203 | 0:3d9c67d97d6f | 540 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 541 | tapCnt--; |
| emh203 | 0:3d9c67d97d6f | 542 | } |
| emh203 | 0:3d9c67d97d6f | 543 | |
| emh203 | 0:3d9c67d97d6f | 544 | /* Advance the state pointer by the decimation factor |
| emh203 | 0:3d9c67d97d6f | 545 | * to process the next group of decimation factor number samples */ |
| emh203 | 0:3d9c67d97d6f | 546 | pState = pState + S->M; |
| emh203 | 0:3d9c67d97d6f | 547 | |
| emh203 | 0:3d9c67d97d6f | 548 | /* Store filter output, smlad returns the values in 2.14 format */ |
| emh203 | 0:3d9c67d97d6f | 549 | /* so downsacle by 15 to get output in 1.15 */ |
| emh203 | 0:3d9c67d97d6f | 550 | *pDst++ = (q15_t) (__SSAT((sum0 >> 15), 16)); |
| emh203 | 0:3d9c67d97d6f | 551 | |
| emh203 | 0:3d9c67d97d6f | 552 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 553 | blkCntN3--; |
| emh203 | 0:3d9c67d97d6f | 554 | } |
| emh203 | 0:3d9c67d97d6f | 555 | |
| emh203 | 0:3d9c67d97d6f | 556 | /* Processing is complete. |
| emh203 | 0:3d9c67d97d6f | 557 | ** Now copy the last numTaps - 1 samples to the satrt of the state buffer. |
| emh203 | 0:3d9c67d97d6f | 558 | ** This prepares the state buffer for the next function call. */ |
| emh203 | 0:3d9c67d97d6f | 559 | |
| emh203 | 0:3d9c67d97d6f | 560 | /* Points to the start of the state buffer */ |
| emh203 | 0:3d9c67d97d6f | 561 | pStateCurnt = S->pState; |
| emh203 | 0:3d9c67d97d6f | 562 | |
| emh203 | 0:3d9c67d97d6f | 563 | i = (numTaps - 1u) >> 2u; |
| emh203 | 0:3d9c67d97d6f | 564 | |
| emh203 | 0:3d9c67d97d6f | 565 | /* copy data */ |
| emh203 | 0:3d9c67d97d6f | 566 | while(i > 0u) |
| emh203 | 0:3d9c67d97d6f | 567 | { |
| emh203 | 0:3d9c67d97d6f | 568 | *pStateCurnt++ = *pState++; |
| emh203 | 0:3d9c67d97d6f | 569 | *pStateCurnt++ = *pState++; |
| emh203 | 0:3d9c67d97d6f | 570 | *pStateCurnt++ = *pState++; |
| emh203 | 0:3d9c67d97d6f | 571 | *pStateCurnt++ = *pState++; |
| emh203 | 0:3d9c67d97d6f | 572 | |
| emh203 | 0:3d9c67d97d6f | 573 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 574 | i--; |
| emh203 | 0:3d9c67d97d6f | 575 | } |
| emh203 | 0:3d9c67d97d6f | 576 | |
| emh203 | 0:3d9c67d97d6f | 577 | i = (numTaps - 1u) % 0x04u; |
| emh203 | 0:3d9c67d97d6f | 578 | |
| emh203 | 0:3d9c67d97d6f | 579 | /* copy data */ |
| emh203 | 0:3d9c67d97d6f | 580 | while(i > 0u) |
| emh203 | 0:3d9c67d97d6f | 581 | { |
| emh203 | 0:3d9c67d97d6f | 582 | *pStateCurnt++ = *pState++; |
| emh203 | 0:3d9c67d97d6f | 583 | |
| emh203 | 0:3d9c67d97d6f | 584 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 585 | i--; |
| emh203 | 0:3d9c67d97d6f | 586 | } |
| emh203 | 0:3d9c67d97d6f | 587 | } |
| emh203 | 0:3d9c67d97d6f | 588 | |
| emh203 | 0:3d9c67d97d6f | 589 | |
| emh203 | 0:3d9c67d97d6f | 590 | #endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */ |
| emh203 | 0:3d9c67d97d6f | 591 | |
| emh203 | 0:3d9c67d97d6f | 592 | #else |
| emh203 | 0:3d9c67d97d6f | 593 | |
| emh203 | 0:3d9c67d97d6f | 594 | |
| emh203 | 0:3d9c67d97d6f | 595 | void arm_fir_decimate_q15( |
| emh203 | 0:3d9c67d97d6f | 596 | const arm_fir_decimate_instance_q15 * S, |
| emh203 | 0:3d9c67d97d6f | 597 | q15_t * pSrc, |
| emh203 | 0:3d9c67d97d6f | 598 | q15_t * pDst, |
| emh203 | 0:3d9c67d97d6f | 599 | uint32_t blockSize) |
| emh203 | 0:3d9c67d97d6f | 600 | { |
| emh203 | 0:3d9c67d97d6f | 601 | q15_t *pState = S->pState; /* State pointer */ |
| emh203 | 0:3d9c67d97d6f | 602 | q15_t *pCoeffs = S->pCoeffs; /* Coefficient pointer */ |
| emh203 | 0:3d9c67d97d6f | 603 | q15_t *pStateCurnt; /* Points to the current sample of the state */ |
| emh203 | 0:3d9c67d97d6f | 604 | q15_t *px; /* Temporary pointer for state buffer */ |
| emh203 | 0:3d9c67d97d6f | 605 | q15_t *pb; /* Temporary pointer coefficient buffer */ |
| emh203 | 0:3d9c67d97d6f | 606 | q31_t x0, c0; /* Temporary variables to hold state and coefficient values */ |
| emh203 | 0:3d9c67d97d6f | 607 | q63_t sum0; /* Accumulators */ |
| emh203 | 0:3d9c67d97d6f | 608 | uint32_t numTaps = S->numTaps; /* Number of taps */ |
| emh203 | 0:3d9c67d97d6f | 609 | uint32_t i, blkCnt, tapCnt, outBlockSize = blockSize / S->M; /* Loop counters */ |
| emh203 | 0:3d9c67d97d6f | 610 | |
| emh203 | 0:3d9c67d97d6f | 611 | |
| emh203 | 0:3d9c67d97d6f | 612 | |
| emh203 | 0:3d9c67d97d6f | 613 | /* Run the below code for Cortex-M0 */ |
| emh203 | 0:3d9c67d97d6f | 614 | |
| emh203 | 0:3d9c67d97d6f | 615 | /* S->pState buffer contains previous frame (numTaps - 1) samples */ |
| emh203 | 0:3d9c67d97d6f | 616 | /* pStateCurnt points to the location where the new input data should be written */ |
| emh203 | 0:3d9c67d97d6f | 617 | pStateCurnt = S->pState + (numTaps - 1u); |
| emh203 | 0:3d9c67d97d6f | 618 | |
| emh203 | 0:3d9c67d97d6f | 619 | /* Total number of output samples to be computed */ |
| emh203 | 0:3d9c67d97d6f | 620 | blkCnt = outBlockSize; |
| emh203 | 0:3d9c67d97d6f | 621 | |
| emh203 | 0:3d9c67d97d6f | 622 | while(blkCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 623 | { |
| emh203 | 0:3d9c67d97d6f | 624 | /* Copy decimation factor number of new input samples into the state buffer */ |
| emh203 | 0:3d9c67d97d6f | 625 | i = S->M; |
| emh203 | 0:3d9c67d97d6f | 626 | |
| emh203 | 0:3d9c67d97d6f | 627 | do |
| emh203 | 0:3d9c67d97d6f | 628 | { |
| emh203 | 0:3d9c67d97d6f | 629 | *pStateCurnt++ = *pSrc++; |
| emh203 | 0:3d9c67d97d6f | 630 | |
| emh203 | 0:3d9c67d97d6f | 631 | } while(--i); |
| emh203 | 0:3d9c67d97d6f | 632 | |
| emh203 | 0:3d9c67d97d6f | 633 | /*Set sum to zero */ |
| emh203 | 0:3d9c67d97d6f | 634 | sum0 = 0; |
| emh203 | 0:3d9c67d97d6f | 635 | |
| emh203 | 0:3d9c67d97d6f | 636 | /* Initialize state pointer */ |
| emh203 | 0:3d9c67d97d6f | 637 | px = pState; |
| emh203 | 0:3d9c67d97d6f | 638 | |
| emh203 | 0:3d9c67d97d6f | 639 | /* Initialize coeff pointer */ |
| emh203 | 0:3d9c67d97d6f | 640 | pb = pCoeffs; |
| emh203 | 0:3d9c67d97d6f | 641 | |
| emh203 | 0:3d9c67d97d6f | 642 | tapCnt = numTaps; |
| emh203 | 0:3d9c67d97d6f | 643 | |
| emh203 | 0:3d9c67d97d6f | 644 | while(tapCnt > 0u) |
| emh203 | 0:3d9c67d97d6f | 645 | { |
| emh203 | 0:3d9c67d97d6f | 646 | /* Read coefficients */ |
| emh203 | 0:3d9c67d97d6f | 647 | c0 = *pb++; |
| emh203 | 0:3d9c67d97d6f | 648 | |
| emh203 | 0:3d9c67d97d6f | 649 | /* Fetch 1 state variable */ |
| emh203 | 0:3d9c67d97d6f | 650 | x0 = *px++; |
| emh203 | 0:3d9c67d97d6f | 651 | |
| emh203 | 0:3d9c67d97d6f | 652 | /* Perform the multiply-accumulate */ |
| emh203 | 0:3d9c67d97d6f | 653 | sum0 += (q31_t) x0 *c0; |
| emh203 | 0:3d9c67d97d6f | 654 | |
| emh203 | 0:3d9c67d97d6f | 655 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 656 | tapCnt--; |
| emh203 | 0:3d9c67d97d6f | 657 | } |
| emh203 | 0:3d9c67d97d6f | 658 | |
| emh203 | 0:3d9c67d97d6f | 659 | /* Advance the state pointer by the decimation factor |
| emh203 | 0:3d9c67d97d6f | 660 | * to process the next group of decimation factor number samples */ |
| emh203 | 0:3d9c67d97d6f | 661 | pState = pState + S->M; |
| emh203 | 0:3d9c67d97d6f | 662 | |
| emh203 | 0:3d9c67d97d6f | 663 | /*Store filter output , smlad will return the values in 2.14 format */ |
| emh203 | 0:3d9c67d97d6f | 664 | /* so downsacle by 15 to get output in 1.15 */ |
| emh203 | 0:3d9c67d97d6f | 665 | *pDst++ = (q15_t) (__SSAT((sum0 >> 15), 16)); |
| emh203 | 0:3d9c67d97d6f | 666 | |
| emh203 | 0:3d9c67d97d6f | 667 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 668 | blkCnt--; |
| emh203 | 0:3d9c67d97d6f | 669 | } |
| emh203 | 0:3d9c67d97d6f | 670 | |
| emh203 | 0:3d9c67d97d6f | 671 | /* Processing is complete. |
| emh203 | 0:3d9c67d97d6f | 672 | ** Now copy the last numTaps - 1 samples to the start of the state buffer. |
| emh203 | 0:3d9c67d97d6f | 673 | ** This prepares the state buffer for the next function call. */ |
| emh203 | 0:3d9c67d97d6f | 674 | |
| emh203 | 0:3d9c67d97d6f | 675 | /* Points to the start of the state buffer */ |
| emh203 | 0:3d9c67d97d6f | 676 | pStateCurnt = S->pState; |
| emh203 | 0:3d9c67d97d6f | 677 | |
| emh203 | 0:3d9c67d97d6f | 678 | i = numTaps - 1u; |
| emh203 | 0:3d9c67d97d6f | 679 | |
| emh203 | 0:3d9c67d97d6f | 680 | /* copy data */ |
| emh203 | 0:3d9c67d97d6f | 681 | while(i > 0u) |
| emh203 | 0:3d9c67d97d6f | 682 | { |
| emh203 | 0:3d9c67d97d6f | 683 | *pStateCurnt++ = *pState++; |
| emh203 | 0:3d9c67d97d6f | 684 | |
| emh203 | 0:3d9c67d97d6f | 685 | /* Decrement the loop counter */ |
| emh203 | 0:3d9c67d97d6f | 686 | i--; |
| emh203 | 0:3d9c67d97d6f | 687 | } |
| emh203 | 0:3d9c67d97d6f | 688 | |
| emh203 | 0:3d9c67d97d6f | 689 | |
| emh203 | 0:3d9c67d97d6f | 690 | } |
| emh203 | 0:3d9c67d97d6f | 691 | #endif /* #ifndef ARM_MATH_CM0_FAMILY */ |
| emh203 | 0:3d9c67d97d6f | 692 | |
| emh203 | 0:3d9c67d97d6f | 693 | |
| emh203 | 0:3d9c67d97d6f | 694 | /** |
| emh203 | 0:3d9c67d97d6f | 695 | * @} end of FIR_decimate group |
| emh203 | 0:3d9c67d97d6f | 696 | */ |
Eli Hughes