Eli Hughes / CMSIS_DSP_401

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_conv_f32.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?

UserRevisionLine numberNew 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_conv_f32.c
emh203 0:3d9c67d97d6f 9 *
emh203 0:3d9c67d97d6f 10 * Description: Convolution of floating-point sequences.
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 * @defgroup Conv Convolution
emh203 0:3d9c67d97d6f 49 *
emh203 0:3d9c67d97d6f 50 * Convolution is a mathematical operation that operates on two finite length vectors to generate a finite length output vector.
emh203 0:3d9c67d97d6f 51 * Convolution is similar to correlation and is frequently used in filtering and data analysis.
emh203 0:3d9c67d97d6f 52 * The CMSIS DSP library contains functions for convolving Q7, Q15, Q31, and floating-point data types.
emh203 0:3d9c67d97d6f 53 * The library also provides fast versions of the Q15 and Q31 functions on Cortex-M4 and Cortex-M3.
emh203 0:3d9c67d97d6f 54 *
emh203 0:3d9c67d97d6f 55 * \par Algorithm
emh203 0:3d9c67d97d6f 56 * Let <code>a[n]</code> and <code>b[n]</code> be sequences of length <code>srcALen</code> and <code>srcBLen</code> samples respectively.
emh203 0:3d9c67d97d6f 57 * Then the convolution
emh203 0:3d9c67d97d6f 58 *
emh203 0:3d9c67d97d6f 59 * <pre>
emh203 0:3d9c67d97d6f 60 * c[n] = a[n] * b[n]
emh203 0:3d9c67d97d6f 61 * </pre>
emh203 0:3d9c67d97d6f 62 *
emh203 0:3d9c67d97d6f 63 * \par
emh203 0:3d9c67d97d6f 64 * is defined as
emh203 0:3d9c67d97d6f 65 * \image html ConvolutionEquation.gif
emh203 0:3d9c67d97d6f 66 * \par
emh203 0:3d9c67d97d6f 67 * Note that <code>c[n]</code> is of length <code>srcALen + srcBLen - 1</code> and is defined over the interval <code>n=0, 1, 2, ..., srcALen + srcBLen - 2</code>.
emh203 0:3d9c67d97d6f 68 * <code>pSrcA</code> points to the first input vector of length <code>srcALen</code> and
emh203 0:3d9c67d97d6f 69 * <code>pSrcB</code> points to the second input vector of length <code>srcBLen</code>.
emh203 0:3d9c67d97d6f 70 * The output result is written to <code>pDst</code> and the calling function must allocate <code>srcALen+srcBLen-1</code> words for the result.
emh203 0:3d9c67d97d6f 71 *
emh203 0:3d9c67d97d6f 72 * \par
emh203 0:3d9c67d97d6f 73 * Conceptually, when two signals <code>a[n]</code> and <code>b[n]</code> are convolved,
emh203 0:3d9c67d97d6f 74 * the signal <code>b[n]</code> slides over <code>a[n]</code>.
emh203 0:3d9c67d97d6f 75 * For each offset \c n, the overlapping portions of a[n] and b[n] are multiplied and summed together.
emh203 0:3d9c67d97d6f 76 *
emh203 0:3d9c67d97d6f 77 * \par
emh203 0:3d9c67d97d6f 78 * Note that convolution is a commutative operation:
emh203 0:3d9c67d97d6f 79 *
emh203 0:3d9c67d97d6f 80 * <pre>
emh203 0:3d9c67d97d6f 81 * a[n] * b[n] = b[n] * a[n].
emh203 0:3d9c67d97d6f 82 * </pre>
emh203 0:3d9c67d97d6f 83 *
emh203 0:3d9c67d97d6f 84 * \par
emh203 0:3d9c67d97d6f 85 * This means that switching the A and B arguments to the convolution functions has no effect.
emh203 0:3d9c67d97d6f 86 *
emh203 0:3d9c67d97d6f 87 * <b>Fixed-Point Behavior</b>
emh203 0:3d9c67d97d6f 88 *
emh203 0:3d9c67d97d6f 89 * \par
emh203 0:3d9c67d97d6f 90 * Convolution requires summing up a large number of intermediate products.
emh203 0:3d9c67d97d6f 91 * As such, the Q7, Q15, and Q31 functions run a risk of overflow and saturation.
emh203 0:3d9c67d97d6f 92 * Refer to the function specific documentation below for further details of the particular algorithm used.
emh203 0:3d9c67d97d6f 93 *
emh203 0:3d9c67d97d6f 94 *
emh203 0:3d9c67d97d6f 95 * <b>Fast Versions</b>
emh203 0:3d9c67d97d6f 96 *
emh203 0:3d9c67d97d6f 97 * \par
emh203 0:3d9c67d97d6f 98 * Fast versions are supported for Q31 and Q15. Cycles for Fast versions are less compared to Q31 and Q15 of conv and the design requires
emh203 0:3d9c67d97d6f 99 * the input signals should be scaled down to avoid intermediate overflows.
emh203 0:3d9c67d97d6f 100 *
emh203 0:3d9c67d97d6f 101 *
emh203 0:3d9c67d97d6f 102 * <b>Opt Versions</b>
emh203 0:3d9c67d97d6f 103 *
emh203 0:3d9c67d97d6f 104 * \par
emh203 0:3d9c67d97d6f 105 * Opt versions are supported for Q15 and Q7. Design uses internal scratch buffer for getting good optimisation.
emh203 0:3d9c67d97d6f 106 * These versions are optimised in cycles and consumes more memory(Scratch memory) compared to Q15 and Q7 versions
emh203 0:3d9c67d97d6f 107 */
emh203 0:3d9c67d97d6f 108
emh203 0:3d9c67d97d6f 109 /**
emh203 0:3d9c67d97d6f 110 * @addtogroup Conv
emh203 0:3d9c67d97d6f 111 * @{
emh203 0:3d9c67d97d6f 112 */
emh203 0:3d9c67d97d6f 113
emh203 0:3d9c67d97d6f 114 /**
emh203 0:3d9c67d97d6f 115 * @brief Convolution of floating-point sequences.
emh203 0:3d9c67d97d6f 116 * @param[in] *pSrcA points to the first input sequence.
emh203 0:3d9c67d97d6f 117 * @param[in] srcALen length of the first input sequence.
emh203 0:3d9c67d97d6f 118 * @param[in] *pSrcB points to the second input sequence.
emh203 0:3d9c67d97d6f 119 * @param[in] srcBLen length of the second input sequence.
emh203 0:3d9c67d97d6f 120 * @param[out] *pDst points to the location where the output result is written. Length srcALen+srcBLen-1.
emh203 0:3d9c67d97d6f 121 * @return none.
emh203 0:3d9c67d97d6f 122 */
emh203 0:3d9c67d97d6f 123
emh203 0:3d9c67d97d6f 124 void arm_conv_f32(
emh203 0:3d9c67d97d6f 125 float32_t * pSrcA,
emh203 0:3d9c67d97d6f 126 uint32_t srcALen,
emh203 0:3d9c67d97d6f 127 float32_t * pSrcB,
emh203 0:3d9c67d97d6f 128 uint32_t srcBLen,
emh203 0:3d9c67d97d6f 129 float32_t * pDst)
emh203 0:3d9c67d97d6f 130 {
emh203 0:3d9c67d97d6f 131
emh203 0:3d9c67d97d6f 132
emh203 0:3d9c67d97d6f 133 #ifndef ARM_MATH_CM0_FAMILY
emh203 0:3d9c67d97d6f 134
emh203 0:3d9c67d97d6f 135 /* Run the below code for Cortex-M4 and Cortex-M3 */
emh203 0:3d9c67d97d6f 136
emh203 0:3d9c67d97d6f 137 float32_t *pIn1; /* inputA pointer */
emh203 0:3d9c67d97d6f 138 float32_t *pIn2; /* inputB pointer */
emh203 0:3d9c67d97d6f 139 float32_t *pOut = pDst; /* output pointer */
emh203 0:3d9c67d97d6f 140 float32_t *px; /* Intermediate inputA pointer */
emh203 0:3d9c67d97d6f 141 float32_t *py; /* Intermediate inputB pointer */
emh203 0:3d9c67d97d6f 142 float32_t *pSrc1, *pSrc2; /* Intermediate pointers */
emh203 0:3d9c67d97d6f 143 float32_t sum, acc0, acc1, acc2, acc3; /* Accumulator */
emh203 0:3d9c67d97d6f 144 float32_t x0, x1, x2, x3, c0; /* Temporary variables to hold state and coefficient values */
emh203 0:3d9c67d97d6f 145 uint32_t j, k, count, blkCnt, blockSize1, blockSize2, blockSize3; /* loop counters */
emh203 0:3d9c67d97d6f 146
emh203 0:3d9c67d97d6f 147 /* The algorithm implementation is based on the lengths of the inputs. */
emh203 0:3d9c67d97d6f 148 /* srcB is always made to slide across srcA. */
emh203 0:3d9c67d97d6f 149 /* So srcBLen is always considered as shorter or equal to srcALen */
emh203 0:3d9c67d97d6f 150 if(srcALen >= srcBLen)
emh203 0:3d9c67d97d6f 151 {
emh203 0:3d9c67d97d6f 152 /* Initialization of inputA pointer */
emh203 0:3d9c67d97d6f 153 pIn1 = pSrcA;
emh203 0:3d9c67d97d6f 154
emh203 0:3d9c67d97d6f 155 /* Initialization of inputB pointer */
emh203 0:3d9c67d97d6f 156 pIn2 = pSrcB;
emh203 0:3d9c67d97d6f 157 }
emh203 0:3d9c67d97d6f 158 else
emh203 0:3d9c67d97d6f 159 {
emh203 0:3d9c67d97d6f 160 /* Initialization of inputA pointer */
emh203 0:3d9c67d97d6f 161 pIn1 = pSrcB;
emh203 0:3d9c67d97d6f 162
emh203 0:3d9c67d97d6f 163 /* Initialization of inputB pointer */
emh203 0:3d9c67d97d6f 164 pIn2 = pSrcA;
emh203 0:3d9c67d97d6f 165
emh203 0:3d9c67d97d6f 166 /* srcBLen is always considered as shorter or equal to srcALen */
emh203 0:3d9c67d97d6f 167 j = srcBLen;
emh203 0:3d9c67d97d6f 168 srcBLen = srcALen;
emh203 0:3d9c67d97d6f 169 srcALen = j;
emh203 0:3d9c67d97d6f 170 }
emh203 0:3d9c67d97d6f 171
emh203 0:3d9c67d97d6f 172 /* conv(x,y) at n = x[n] * y[0] + x[n-1] * y[1] + x[n-2] * y[2] + ...+ x[n-N+1] * y[N -1] */
emh203 0:3d9c67d97d6f 173 /* The function is internally
emh203 0:3d9c67d97d6f 174 * divided into three stages according to the number of multiplications that has to be
emh203 0:3d9c67d97d6f 175 * taken place between inputA samples and inputB samples. In the first stage of the
emh203 0:3d9c67d97d6f 176 * algorithm, the multiplications increase by one for every iteration.
emh203 0:3d9c67d97d6f 177 * In the second stage of the algorithm, srcBLen number of multiplications are done.
emh203 0:3d9c67d97d6f 178 * In the third stage of the algorithm, the multiplications decrease by one
emh203 0:3d9c67d97d6f 179 * for every iteration. */
emh203 0:3d9c67d97d6f 180
emh203 0:3d9c67d97d6f 181 /* The algorithm is implemented in three stages.
emh203 0:3d9c67d97d6f 182 The loop counters of each stage is initiated here. */
emh203 0:3d9c67d97d6f 183 blockSize1 = srcBLen - 1u;
emh203 0:3d9c67d97d6f 184 blockSize2 = srcALen - (srcBLen - 1u);
emh203 0:3d9c67d97d6f 185 blockSize3 = blockSize1;
emh203 0:3d9c67d97d6f 186
emh203 0:3d9c67d97d6f 187 /* --------------------------
emh203 0:3d9c67d97d6f 188 * initializations of stage1
emh203 0:3d9c67d97d6f 189 * -------------------------*/
emh203 0:3d9c67d97d6f 190
emh203 0:3d9c67d97d6f 191 /* sum = x[0] * y[0]
emh203 0:3d9c67d97d6f 192 * sum = x[0] * y[1] + x[1] * y[0]
emh203 0:3d9c67d97d6f 193 * ....
emh203 0:3d9c67d97d6f 194 * sum = x[0] * y[srcBlen - 1] + x[1] * y[srcBlen - 2] +...+ x[srcBLen - 1] * y[0]
emh203 0:3d9c67d97d6f 195 */
emh203 0:3d9c67d97d6f 196
emh203 0:3d9c67d97d6f 197 /* In this stage the MAC operations are increased by 1 for every iteration.
emh203 0:3d9c67d97d6f 198 The count variable holds the number of MAC operations performed */
emh203 0:3d9c67d97d6f 199 count = 1u;
emh203 0:3d9c67d97d6f 200
emh203 0:3d9c67d97d6f 201 /* Working pointer of inputA */
emh203 0:3d9c67d97d6f 202 px = pIn1;
emh203 0:3d9c67d97d6f 203
emh203 0:3d9c67d97d6f 204 /* Working pointer of inputB */
emh203 0:3d9c67d97d6f 205 py = pIn2;
emh203 0:3d9c67d97d6f 206
emh203 0:3d9c67d97d6f 207
emh203 0:3d9c67d97d6f 208 /* ------------------------
emh203 0:3d9c67d97d6f 209 * Stage1 process
emh203 0:3d9c67d97d6f 210 * ----------------------*/
emh203 0:3d9c67d97d6f 211
emh203 0:3d9c67d97d6f 212 /* The first stage starts here */
emh203 0:3d9c67d97d6f 213 while(blockSize1 > 0u)
emh203 0:3d9c67d97d6f 214 {
emh203 0:3d9c67d97d6f 215 /* Accumulator is made zero for every iteration */
emh203 0:3d9c67d97d6f 216 sum = 0.0f;
emh203 0:3d9c67d97d6f 217
emh203 0:3d9c67d97d6f 218 /* Apply loop unrolling and compute 4 MACs simultaneously. */
emh203 0:3d9c67d97d6f 219 k = count >> 2u;
emh203 0:3d9c67d97d6f 220
emh203 0:3d9c67d97d6f 221 /* First part of the processing with loop unrolling. Compute 4 MACs at a time.
emh203 0:3d9c67d97d6f 222 ** a second loop below computes MACs for the remaining 1 to 3 samples. */
emh203 0:3d9c67d97d6f 223 while(k > 0u)
emh203 0:3d9c67d97d6f 224 {
emh203 0:3d9c67d97d6f 225 /* x[0] * y[srcBLen - 1] */
emh203 0:3d9c67d97d6f 226 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 227
emh203 0:3d9c67d97d6f 228 /* x[1] * y[srcBLen - 2] */
emh203 0:3d9c67d97d6f 229 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 230
emh203 0:3d9c67d97d6f 231 /* x[2] * y[srcBLen - 3] */
emh203 0:3d9c67d97d6f 232 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 233
emh203 0:3d9c67d97d6f 234 /* x[3] * y[srcBLen - 4] */
emh203 0:3d9c67d97d6f 235 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 236
emh203 0:3d9c67d97d6f 237 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 238 k--;
emh203 0:3d9c67d97d6f 239 }
emh203 0:3d9c67d97d6f 240
emh203 0:3d9c67d97d6f 241 /* If the count is not a multiple of 4, compute any remaining MACs here.
emh203 0:3d9c67d97d6f 242 ** No loop unrolling is used. */
emh203 0:3d9c67d97d6f 243 k = count % 0x4u;
emh203 0:3d9c67d97d6f 244
emh203 0:3d9c67d97d6f 245 while(k > 0u)
emh203 0:3d9c67d97d6f 246 {
emh203 0:3d9c67d97d6f 247 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 248 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 249
emh203 0:3d9c67d97d6f 250 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 251 k--;
emh203 0:3d9c67d97d6f 252 }
emh203 0:3d9c67d97d6f 253
emh203 0:3d9c67d97d6f 254 /* Store the result in the accumulator in the destination buffer. */
emh203 0:3d9c67d97d6f 255 *pOut++ = sum;
emh203 0:3d9c67d97d6f 256
emh203 0:3d9c67d97d6f 257 /* Update the inputA and inputB pointers for next MAC calculation */
emh203 0:3d9c67d97d6f 258 py = pIn2 + count;
emh203 0:3d9c67d97d6f 259 px = pIn1;
emh203 0:3d9c67d97d6f 260
emh203 0:3d9c67d97d6f 261 /* Increment the MAC count */
emh203 0:3d9c67d97d6f 262 count++;
emh203 0:3d9c67d97d6f 263
emh203 0:3d9c67d97d6f 264 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 265 blockSize1--;
emh203 0:3d9c67d97d6f 266 }
emh203 0:3d9c67d97d6f 267
emh203 0:3d9c67d97d6f 268 /* --------------------------
emh203 0:3d9c67d97d6f 269 * Initializations of stage2
emh203 0:3d9c67d97d6f 270 * ------------------------*/
emh203 0:3d9c67d97d6f 271
emh203 0:3d9c67d97d6f 272 /* sum = x[0] * y[srcBLen-1] + x[1] * y[srcBLen-2] +...+ x[srcBLen-1] * y[0]
emh203 0:3d9c67d97d6f 273 * sum = x[1] * y[srcBLen-1] + x[2] * y[srcBLen-2] +...+ x[srcBLen] * y[0]
emh203 0:3d9c67d97d6f 274 * ....
emh203 0:3d9c67d97d6f 275 * sum = x[srcALen-srcBLen-2] * y[srcBLen-1] + x[srcALen] * y[srcBLen-2] +...+ x[srcALen-1] * y[0]
emh203 0:3d9c67d97d6f 276 */
emh203 0:3d9c67d97d6f 277
emh203 0:3d9c67d97d6f 278 /* Working pointer of inputA */
emh203 0:3d9c67d97d6f 279 px = pIn1;
emh203 0:3d9c67d97d6f 280
emh203 0:3d9c67d97d6f 281 /* Working pointer of inputB */
emh203 0:3d9c67d97d6f 282 pSrc2 = pIn2 + (srcBLen - 1u);
emh203 0:3d9c67d97d6f 283 py = pSrc2;
emh203 0:3d9c67d97d6f 284
emh203 0:3d9c67d97d6f 285 /* count is index by which the pointer pIn1 to be incremented */
emh203 0:3d9c67d97d6f 286 count = 0u;
emh203 0:3d9c67d97d6f 287
emh203 0:3d9c67d97d6f 288 /* -------------------
emh203 0:3d9c67d97d6f 289 * Stage2 process
emh203 0:3d9c67d97d6f 290 * ------------------*/
emh203 0:3d9c67d97d6f 291
emh203 0:3d9c67d97d6f 292 /* Stage2 depends on srcBLen as in this stage srcBLen number of MACS are performed.
emh203 0:3d9c67d97d6f 293 * So, to loop unroll over blockSize2,
emh203 0:3d9c67d97d6f 294 * srcBLen should be greater than or equal to 4 */
emh203 0:3d9c67d97d6f 295 if(srcBLen >= 4u)
emh203 0:3d9c67d97d6f 296 {
emh203 0:3d9c67d97d6f 297 /* Loop unroll over blockSize2, by 4 */
emh203 0:3d9c67d97d6f 298 blkCnt = blockSize2 >> 2u;
emh203 0:3d9c67d97d6f 299
emh203 0:3d9c67d97d6f 300 while(blkCnt > 0u)
emh203 0:3d9c67d97d6f 301 {
emh203 0:3d9c67d97d6f 302 /* Set all accumulators to zero */
emh203 0:3d9c67d97d6f 303 acc0 = 0.0f;
emh203 0:3d9c67d97d6f 304 acc1 = 0.0f;
emh203 0:3d9c67d97d6f 305 acc2 = 0.0f;
emh203 0:3d9c67d97d6f 306 acc3 = 0.0f;
emh203 0:3d9c67d97d6f 307
emh203 0:3d9c67d97d6f 308 /* read x[0], x[1], x[2] samples */
emh203 0:3d9c67d97d6f 309 x0 = *(px++);
emh203 0:3d9c67d97d6f 310 x1 = *(px++);
emh203 0:3d9c67d97d6f 311 x2 = *(px++);
emh203 0:3d9c67d97d6f 312
emh203 0:3d9c67d97d6f 313 /* Apply loop unrolling and compute 4 MACs simultaneously. */
emh203 0:3d9c67d97d6f 314 k = srcBLen >> 2u;
emh203 0:3d9c67d97d6f 315
emh203 0:3d9c67d97d6f 316 /* First part of the processing with loop unrolling. Compute 4 MACs at a time.
emh203 0:3d9c67d97d6f 317 ** a second loop below computes MACs for the remaining 1 to 3 samples. */
emh203 0:3d9c67d97d6f 318 do
emh203 0:3d9c67d97d6f 319 {
emh203 0:3d9c67d97d6f 320 /* Read y[srcBLen - 1] sample */
emh203 0:3d9c67d97d6f 321 c0 = *(py--);
emh203 0:3d9c67d97d6f 322
emh203 0:3d9c67d97d6f 323 /* Read x[3] sample */
emh203 0:3d9c67d97d6f 324 x3 = *(px);
emh203 0:3d9c67d97d6f 325
emh203 0:3d9c67d97d6f 326 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 327 /* acc0 += x[0] * y[srcBLen - 1] */
emh203 0:3d9c67d97d6f 328 acc0 += x0 * c0;
emh203 0:3d9c67d97d6f 329
emh203 0:3d9c67d97d6f 330 /* acc1 += x[1] * y[srcBLen - 1] */
emh203 0:3d9c67d97d6f 331 acc1 += x1 * c0;
emh203 0:3d9c67d97d6f 332
emh203 0:3d9c67d97d6f 333 /* acc2 += x[2] * y[srcBLen - 1] */
emh203 0:3d9c67d97d6f 334 acc2 += x2 * c0;
emh203 0:3d9c67d97d6f 335
emh203 0:3d9c67d97d6f 336 /* acc3 += x[3] * y[srcBLen - 1] */
emh203 0:3d9c67d97d6f 337 acc3 += x3 * c0;
emh203 0:3d9c67d97d6f 338
emh203 0:3d9c67d97d6f 339 /* Read y[srcBLen - 2] sample */
emh203 0:3d9c67d97d6f 340 c0 = *(py--);
emh203 0:3d9c67d97d6f 341
emh203 0:3d9c67d97d6f 342 /* Read x[4] sample */
emh203 0:3d9c67d97d6f 343 x0 = *(px + 1u);
emh203 0:3d9c67d97d6f 344
emh203 0:3d9c67d97d6f 345 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 346 /* acc0 += x[1] * y[srcBLen - 2] */
emh203 0:3d9c67d97d6f 347 acc0 += x1 * c0;
emh203 0:3d9c67d97d6f 348 /* acc1 += x[2] * y[srcBLen - 2] */
emh203 0:3d9c67d97d6f 349 acc1 += x2 * c0;
emh203 0:3d9c67d97d6f 350 /* acc2 += x[3] * y[srcBLen - 2] */
emh203 0:3d9c67d97d6f 351 acc2 += x3 * c0;
emh203 0:3d9c67d97d6f 352 /* acc3 += x[4] * y[srcBLen - 2] */
emh203 0:3d9c67d97d6f 353 acc3 += x0 * c0;
emh203 0:3d9c67d97d6f 354
emh203 0:3d9c67d97d6f 355 /* Read y[srcBLen - 3] sample */
emh203 0:3d9c67d97d6f 356 c0 = *(py--);
emh203 0:3d9c67d97d6f 357
emh203 0:3d9c67d97d6f 358 /* Read x[5] sample */
emh203 0:3d9c67d97d6f 359 x1 = *(px + 2u);
emh203 0:3d9c67d97d6f 360
emh203 0:3d9c67d97d6f 361 /* Perform the multiply-accumulates */
emh203 0:3d9c67d97d6f 362 /* acc0 += x[2] * y[srcBLen - 3] */
emh203 0:3d9c67d97d6f 363 acc0 += x2 * c0;
emh203 0:3d9c67d97d6f 364 /* acc1 += x[3] * y[srcBLen - 2] */
emh203 0:3d9c67d97d6f 365 acc1 += x3 * c0;
emh203 0:3d9c67d97d6f 366 /* acc2 += x[4] * y[srcBLen - 2] */
emh203 0:3d9c67d97d6f 367 acc2 += x0 * c0;
emh203 0:3d9c67d97d6f 368 /* acc3 += x[5] * y[srcBLen - 2] */
emh203 0:3d9c67d97d6f 369 acc3 += x1 * c0;
emh203 0:3d9c67d97d6f 370
emh203 0:3d9c67d97d6f 371 /* Read y[srcBLen - 4] sample */
emh203 0:3d9c67d97d6f 372 c0 = *(py--);
emh203 0:3d9c67d97d6f 373
emh203 0:3d9c67d97d6f 374 /* Read x[6] sample */
emh203 0:3d9c67d97d6f 375 x2 = *(px + 3u);
emh203 0:3d9c67d97d6f 376 px += 4u;
emh203 0:3d9c67d97d6f 377
emh203 0:3d9c67d97d6f 378 /* Perform the multiply-accumulates */
emh203 0:3d9c67d97d6f 379 /* acc0 += x[3] * y[srcBLen - 4] */
emh203 0:3d9c67d97d6f 380 acc0 += x3 * c0;
emh203 0:3d9c67d97d6f 381 /* acc1 += x[4] * y[srcBLen - 4] */
emh203 0:3d9c67d97d6f 382 acc1 += x0 * c0;
emh203 0:3d9c67d97d6f 383 /* acc2 += x[5] * y[srcBLen - 4] */
emh203 0:3d9c67d97d6f 384 acc2 += x1 * c0;
emh203 0:3d9c67d97d6f 385 /* acc3 += x[6] * y[srcBLen - 4] */
emh203 0:3d9c67d97d6f 386 acc3 += x2 * c0;
emh203 0:3d9c67d97d6f 387
emh203 0:3d9c67d97d6f 388
emh203 0:3d9c67d97d6f 389 } while(--k);
emh203 0:3d9c67d97d6f 390
emh203 0:3d9c67d97d6f 391 /* If the srcBLen is not a multiple of 4, compute any remaining MACs here.
emh203 0:3d9c67d97d6f 392 ** No loop unrolling is used. */
emh203 0:3d9c67d97d6f 393 k = srcBLen % 0x4u;
emh203 0:3d9c67d97d6f 394
emh203 0:3d9c67d97d6f 395 while(k > 0u)
emh203 0:3d9c67d97d6f 396 {
emh203 0:3d9c67d97d6f 397 /* Read y[srcBLen - 5] sample */
emh203 0:3d9c67d97d6f 398 c0 = *(py--);
emh203 0:3d9c67d97d6f 399
emh203 0:3d9c67d97d6f 400 /* Read x[7] sample */
emh203 0:3d9c67d97d6f 401 x3 = *(px++);
emh203 0:3d9c67d97d6f 402
emh203 0:3d9c67d97d6f 403 /* Perform the multiply-accumulates */
emh203 0:3d9c67d97d6f 404 /* acc0 += x[4] * y[srcBLen - 5] */
emh203 0:3d9c67d97d6f 405 acc0 += x0 * c0;
emh203 0:3d9c67d97d6f 406 /* acc1 += x[5] * y[srcBLen - 5] */
emh203 0:3d9c67d97d6f 407 acc1 += x1 * c0;
emh203 0:3d9c67d97d6f 408 /* acc2 += x[6] * y[srcBLen - 5] */
emh203 0:3d9c67d97d6f 409 acc2 += x2 * c0;
emh203 0:3d9c67d97d6f 410 /* acc3 += x[7] * y[srcBLen - 5] */
emh203 0:3d9c67d97d6f 411 acc3 += x3 * c0;
emh203 0:3d9c67d97d6f 412
emh203 0:3d9c67d97d6f 413 /* Reuse the present samples for the next MAC */
emh203 0:3d9c67d97d6f 414 x0 = x1;
emh203 0:3d9c67d97d6f 415 x1 = x2;
emh203 0:3d9c67d97d6f 416 x2 = x3;
emh203 0:3d9c67d97d6f 417
emh203 0:3d9c67d97d6f 418 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 419 k--;
emh203 0:3d9c67d97d6f 420 }
emh203 0:3d9c67d97d6f 421
emh203 0:3d9c67d97d6f 422 /* Store the result in the accumulator in the destination buffer. */
emh203 0:3d9c67d97d6f 423 *pOut++ = acc0;
emh203 0:3d9c67d97d6f 424 *pOut++ = acc1;
emh203 0:3d9c67d97d6f 425 *pOut++ = acc2;
emh203 0:3d9c67d97d6f 426 *pOut++ = acc3;
emh203 0:3d9c67d97d6f 427
emh203 0:3d9c67d97d6f 428 /* Increment the pointer pIn1 index, count by 4 */
emh203 0:3d9c67d97d6f 429 count += 4u;
emh203 0:3d9c67d97d6f 430
emh203 0:3d9c67d97d6f 431 /* Update the inputA and inputB pointers for next MAC calculation */
emh203 0:3d9c67d97d6f 432 px = pIn1 + count;
emh203 0:3d9c67d97d6f 433 py = pSrc2;
emh203 0:3d9c67d97d6f 434
emh203 0:3d9c67d97d6f 435
emh203 0:3d9c67d97d6f 436 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 437 blkCnt--;
emh203 0:3d9c67d97d6f 438 }
emh203 0:3d9c67d97d6f 439
emh203 0:3d9c67d97d6f 440
emh203 0:3d9c67d97d6f 441 /* If the blockSize2 is not a multiple of 4, compute any remaining output samples here.
emh203 0:3d9c67d97d6f 442 ** No loop unrolling is used. */
emh203 0:3d9c67d97d6f 443 blkCnt = blockSize2 % 0x4u;
emh203 0:3d9c67d97d6f 444
emh203 0:3d9c67d97d6f 445 while(blkCnt > 0u)
emh203 0:3d9c67d97d6f 446 {
emh203 0:3d9c67d97d6f 447 /* Accumulator is made zero for every iteration */
emh203 0:3d9c67d97d6f 448 sum = 0.0f;
emh203 0:3d9c67d97d6f 449
emh203 0:3d9c67d97d6f 450 /* Apply loop unrolling and compute 4 MACs simultaneously. */
emh203 0:3d9c67d97d6f 451 k = srcBLen >> 2u;
emh203 0:3d9c67d97d6f 452
emh203 0:3d9c67d97d6f 453 /* First part of the processing with loop unrolling. Compute 4 MACs at a time.
emh203 0:3d9c67d97d6f 454 ** a second loop below computes MACs for the remaining 1 to 3 samples. */
emh203 0:3d9c67d97d6f 455 while(k > 0u)
emh203 0:3d9c67d97d6f 456 {
emh203 0:3d9c67d97d6f 457 /* Perform the multiply-accumulates */
emh203 0:3d9c67d97d6f 458 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 459 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 460 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 461 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 462
emh203 0:3d9c67d97d6f 463 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 464 k--;
emh203 0:3d9c67d97d6f 465 }
emh203 0:3d9c67d97d6f 466
emh203 0:3d9c67d97d6f 467 /* If the srcBLen is not a multiple of 4, compute any remaining MACs here.
emh203 0:3d9c67d97d6f 468 ** No loop unrolling is used. */
emh203 0:3d9c67d97d6f 469 k = srcBLen % 0x4u;
emh203 0:3d9c67d97d6f 470
emh203 0:3d9c67d97d6f 471 while(k > 0u)
emh203 0:3d9c67d97d6f 472 {
emh203 0:3d9c67d97d6f 473 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 474 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 475
emh203 0:3d9c67d97d6f 476 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 477 k--;
emh203 0:3d9c67d97d6f 478 }
emh203 0:3d9c67d97d6f 479
emh203 0:3d9c67d97d6f 480 /* Store the result in the accumulator in the destination buffer. */
emh203 0:3d9c67d97d6f 481 *pOut++ = sum;
emh203 0:3d9c67d97d6f 482
emh203 0:3d9c67d97d6f 483 /* Increment the MAC count */
emh203 0:3d9c67d97d6f 484 count++;
emh203 0:3d9c67d97d6f 485
emh203 0:3d9c67d97d6f 486 /* Update the inputA and inputB pointers for next MAC calculation */
emh203 0:3d9c67d97d6f 487 px = pIn1 + count;
emh203 0:3d9c67d97d6f 488 py = pSrc2;
emh203 0:3d9c67d97d6f 489
emh203 0:3d9c67d97d6f 490 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 491 blkCnt--;
emh203 0:3d9c67d97d6f 492 }
emh203 0:3d9c67d97d6f 493 }
emh203 0:3d9c67d97d6f 494 else
emh203 0:3d9c67d97d6f 495 {
emh203 0:3d9c67d97d6f 496 /* If the srcBLen is not a multiple of 4,
emh203 0:3d9c67d97d6f 497 * the blockSize2 loop cannot be unrolled by 4 */
emh203 0:3d9c67d97d6f 498 blkCnt = blockSize2;
emh203 0:3d9c67d97d6f 499
emh203 0:3d9c67d97d6f 500 while(blkCnt > 0u)
emh203 0:3d9c67d97d6f 501 {
emh203 0:3d9c67d97d6f 502 /* Accumulator is made zero for every iteration */
emh203 0:3d9c67d97d6f 503 sum = 0.0f;
emh203 0:3d9c67d97d6f 504
emh203 0:3d9c67d97d6f 505 /* srcBLen number of MACS should be performed */
emh203 0:3d9c67d97d6f 506 k = srcBLen;
emh203 0:3d9c67d97d6f 507
emh203 0:3d9c67d97d6f 508 while(k > 0u)
emh203 0:3d9c67d97d6f 509 {
emh203 0:3d9c67d97d6f 510 /* Perform the multiply-accumulate */
emh203 0:3d9c67d97d6f 511 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 512
emh203 0:3d9c67d97d6f 513 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 514 k--;
emh203 0:3d9c67d97d6f 515 }
emh203 0:3d9c67d97d6f 516
emh203 0:3d9c67d97d6f 517 /* Store the result in the accumulator in the destination buffer. */
emh203 0:3d9c67d97d6f 518 *pOut++ = sum;
emh203 0:3d9c67d97d6f 519
emh203 0:3d9c67d97d6f 520 /* Increment the MAC count */
emh203 0:3d9c67d97d6f 521 count++;
emh203 0:3d9c67d97d6f 522
emh203 0:3d9c67d97d6f 523 /* Update the inputA and inputB pointers for next MAC calculation */
emh203 0:3d9c67d97d6f 524 px = pIn1 + count;
emh203 0:3d9c67d97d6f 525 py = pSrc2;
emh203 0:3d9c67d97d6f 526
emh203 0:3d9c67d97d6f 527 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 528 blkCnt--;
emh203 0:3d9c67d97d6f 529 }
emh203 0:3d9c67d97d6f 530 }
emh203 0:3d9c67d97d6f 531
emh203 0:3d9c67d97d6f 532
emh203 0:3d9c67d97d6f 533 /* --------------------------
emh203 0:3d9c67d97d6f 534 * Initializations of stage3
emh203 0:3d9c67d97d6f 535 * -------------------------*/
emh203 0:3d9c67d97d6f 536
emh203 0:3d9c67d97d6f 537 /* sum += x[srcALen-srcBLen+1] * y[srcBLen-1] + x[srcALen-srcBLen+2] * y[srcBLen-2] +...+ x[srcALen-1] * y[1]
emh203 0:3d9c67d97d6f 538 * sum += x[srcALen-srcBLen+2] * y[srcBLen-1] + x[srcALen-srcBLen+3] * y[srcBLen-2] +...+ x[srcALen-1] * y[2]
emh203 0:3d9c67d97d6f 539 * ....
emh203 0:3d9c67d97d6f 540 * sum += x[srcALen-2] * y[srcBLen-1] + x[srcALen-1] * y[srcBLen-2]
emh203 0:3d9c67d97d6f 541 * sum += x[srcALen-1] * y[srcBLen-1]
emh203 0:3d9c67d97d6f 542 */
emh203 0:3d9c67d97d6f 543
emh203 0:3d9c67d97d6f 544 /* In this stage the MAC operations are decreased by 1 for every iteration.
emh203 0:3d9c67d97d6f 545 The blockSize3 variable holds the number of MAC operations performed */
emh203 0:3d9c67d97d6f 546
emh203 0:3d9c67d97d6f 547 /* Working pointer of inputA */
emh203 0:3d9c67d97d6f 548 pSrc1 = (pIn1 + srcALen) - (srcBLen - 1u);
emh203 0:3d9c67d97d6f 549 px = pSrc1;
emh203 0:3d9c67d97d6f 550
emh203 0:3d9c67d97d6f 551 /* Working pointer of inputB */
emh203 0:3d9c67d97d6f 552 pSrc2 = pIn2 + (srcBLen - 1u);
emh203 0:3d9c67d97d6f 553 py = pSrc2;
emh203 0:3d9c67d97d6f 554
emh203 0:3d9c67d97d6f 555 /* -------------------
emh203 0:3d9c67d97d6f 556 * Stage3 process
emh203 0:3d9c67d97d6f 557 * ------------------*/
emh203 0:3d9c67d97d6f 558
emh203 0:3d9c67d97d6f 559 while(blockSize3 > 0u)
emh203 0:3d9c67d97d6f 560 {
emh203 0:3d9c67d97d6f 561 /* Accumulator is made zero for every iteration */
emh203 0:3d9c67d97d6f 562 sum = 0.0f;
emh203 0:3d9c67d97d6f 563
emh203 0:3d9c67d97d6f 564 /* Apply loop unrolling and compute 4 MACs simultaneously. */
emh203 0:3d9c67d97d6f 565 k = blockSize3 >> 2u;
emh203 0:3d9c67d97d6f 566
emh203 0:3d9c67d97d6f 567 /* First part of the processing with loop unrolling. Compute 4 MACs at a time.
emh203 0:3d9c67d97d6f 568 ** a second loop below computes MACs for the remaining 1 to 3 samples. */
emh203 0:3d9c67d97d6f 569 while(k > 0u)
emh203 0:3d9c67d97d6f 570 {
emh203 0:3d9c67d97d6f 571 /* sum += x[srcALen - srcBLen + 1] * y[srcBLen - 1] */
emh203 0:3d9c67d97d6f 572 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 573
emh203 0:3d9c67d97d6f 574 /* sum += x[srcALen - srcBLen + 2] * y[srcBLen - 2] */
emh203 0:3d9c67d97d6f 575 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 576
emh203 0:3d9c67d97d6f 577 /* sum += x[srcALen - srcBLen + 3] * y[srcBLen - 3] */
emh203 0:3d9c67d97d6f 578 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 579
emh203 0:3d9c67d97d6f 580 /* sum += x[srcALen - srcBLen + 4] * y[srcBLen - 4] */
emh203 0:3d9c67d97d6f 581 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 582
emh203 0:3d9c67d97d6f 583 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 584 k--;
emh203 0:3d9c67d97d6f 585 }
emh203 0:3d9c67d97d6f 586
emh203 0:3d9c67d97d6f 587 /* If the blockSize3 is not a multiple of 4, compute any remaining MACs here.
emh203 0:3d9c67d97d6f 588 ** No loop unrolling is used. */
emh203 0:3d9c67d97d6f 589 k = blockSize3 % 0x4u;
emh203 0:3d9c67d97d6f 590
emh203 0:3d9c67d97d6f 591 while(k > 0u)
emh203 0:3d9c67d97d6f 592 {
emh203 0:3d9c67d97d6f 593 /* Perform the multiply-accumulates */
emh203 0:3d9c67d97d6f 594 /* sum += x[srcALen-1] * y[srcBLen-1] */
emh203 0:3d9c67d97d6f 595 sum += *px++ * *py--;
emh203 0:3d9c67d97d6f 596
emh203 0:3d9c67d97d6f 597 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 598 k--;
emh203 0:3d9c67d97d6f 599 }
emh203 0:3d9c67d97d6f 600
emh203 0:3d9c67d97d6f 601 /* Store the result in the accumulator in the destination buffer. */
emh203 0:3d9c67d97d6f 602 *pOut++ = sum;
emh203 0:3d9c67d97d6f 603
emh203 0:3d9c67d97d6f 604 /* Update the inputA and inputB pointers for next MAC calculation */
emh203 0:3d9c67d97d6f 605 px = ++pSrc1;
emh203 0:3d9c67d97d6f 606 py = pSrc2;
emh203 0:3d9c67d97d6f 607
emh203 0:3d9c67d97d6f 608 /* Decrement the loop counter */
emh203 0:3d9c67d97d6f 609 blockSize3--;
emh203 0:3d9c67d97d6f 610 }
emh203 0:3d9c67d97d6f 611
emh203 0:3d9c67d97d6f 612 #else
emh203 0:3d9c67d97d6f 613
emh203 0:3d9c67d97d6f 614 /* Run the below code for Cortex-M0 */
emh203 0:3d9c67d97d6f 615
emh203 0:3d9c67d97d6f 616 float32_t *pIn1 = pSrcA; /* inputA pointer */
emh203 0:3d9c67d97d6f 617 float32_t *pIn2 = pSrcB; /* inputB pointer */
emh203 0:3d9c67d97d6f 618 float32_t sum; /* Accumulator */
emh203 0:3d9c67d97d6f 619 uint32_t i, j; /* loop counters */
emh203 0:3d9c67d97d6f 620
emh203 0:3d9c67d97d6f 621 /* Loop to calculate convolution for output length number of times */
emh203 0:3d9c67d97d6f 622 for (i = 0u; i < ((srcALen + srcBLen) - 1u); i++)
emh203 0:3d9c67d97d6f 623 {
emh203 0:3d9c67d97d6f 624 /* Initialize sum with zero to carry out MAC operations */
emh203 0:3d9c67d97d6f 625 sum = 0.0f;
emh203 0:3d9c67d97d6f 626
emh203 0:3d9c67d97d6f 627 /* Loop to perform MAC operations according to convolution equation */
emh203 0:3d9c67d97d6f 628 for (j = 0u; j <= i; j++)
emh203 0:3d9c67d97d6f 629 {
emh203 0:3d9c67d97d6f 630 /* Check the array limitations */
emh203 0:3d9c67d97d6f 631 if((((i - j) < srcBLen) && (j < srcALen)))
emh203 0:3d9c67d97d6f 632 {
emh203 0:3d9c67d97d6f 633 /* z[i] += x[i-j] * y[j] */
emh203 0:3d9c67d97d6f 634 sum += pIn1[j] * pIn2[i - j];
emh203 0:3d9c67d97d6f 635 }
emh203 0:3d9c67d97d6f 636 }
emh203 0:3d9c67d97d6f 637 /* Store the output in the destination buffer */
emh203 0:3d9c67d97d6f 638 pDst[i] = sum;
emh203 0:3d9c67d97d6f 639 }
emh203 0:3d9c67d97d6f 640
emh203 0:3d9c67d97d6f 641 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
emh203 0:3d9c67d97d6f 642
emh203 0:3d9c67d97d6f 643 }
emh203 0:3d9c67d97d6f 644
emh203 0:3d9c67d97d6f 645 /**
emh203 0:3d9c67d97d6f 646 * @} end of Conv group
emh203 0:3d9c67d97d6f 647 */