Renesas / mbed-dsp-neon

Fork of mbed-dsp. CMSIS-DSP library of supporting NEON

Dependents:   mbed-os-example-cmsis_dsp_neon

Fork of mbed-dsp by mbed official

Information

Japanese version is available in lower part of this page.
このページの後半に日本語版が用意されています．

CMSIS-DSP of supporting NEON

What is this ?

A library for CMSIS-DSP of supporting NEON.
We supported the NEON to CMSIS-DSP Ver1.4.3(CMSIS V4.1) that ARM supplied, has achieved the processing speed improvement.
If you use the mbed-dsp library, you can use to replace this library.
CMSIS-DSP of supporting NEON is provied as a library.

Library Creation environment

CMSIS-DSP library of supporting NEON was created by the following environment.

  • Compiler
    ARMCC Version 5.03
  • Compile option switch[C Compiler]
   -DARM_MATH_MATRIX_CHECK -DARM_MATH_ROUNDING -O3 -Otime --cpu=Cortex-A9 --littleend --arm 
   --apcs=/interwork --no_unaligned_access --fpu=vfpv3_fp16 --fpmode=fast --apcs=/hardfp 
   --vectorize --asm
  • Compile option switch[Assembler]
   --cpreproc --cpu=Cortex-A9 --littleend --arm --apcs=/interwork --no_unaligned_access 
   --fpu=vfpv3_fp16 --fpmode=fast --apcs=/hardfp


Effects of NEON support

In the data which passes to each function, large size will be expected more effective than small size.
Also if the data is a multiple of 16, effect will be expected in every function in the CMSIS-DSP.


NEON対応CMSIS-DSP

概要

NEON対応したCMSIS-DSPのライブラリです。
ARM社提供のCMSIS-DSP Ver1.4.3(CMSIS V4.1)をターゲットにNEON対応を行ない、処理速度向上を実現しております。
mbed-dspライブラリを使用している場合は、本ライブラリに置き換えて使用することができます。
NEON対応したCMSIS-DSPはライブラリで提供します。

ライブラリ作成環境

NEON対応CMSIS-DSPライブラリは、以下の環境で作成しています。

  • コンパイラ
    ARMCC Version 5.03
  • コンパイルオプションスイッチ[C Compiler]
   -DARM_MATH_MATRIX_CHECK -DARM_MATH_ROUNDING -O3 -Otime --cpu=Cortex-A9 --littleend --arm 
   --apcs=/interwork --no_unaligned_access --fpu=vfpv3_fp16 --fpmode=fast --apcs=/hardfp 
   --vectorize --asm
  • コンパイルオプションスイッチ[Assembler]
   --cpreproc --cpu=Cortex-A9 --littleend --arm --apcs=/interwork --no_unaligned_access 
   --fpu=vfpv3_fp16 --fpmode=fast --apcs=/hardfp


NEON対応による効果について

CMSIS-DSP内の各関数へ渡すデータは、小さいサイズよりも大きいサイズの方が効果が見込めます。
また、16の倍数のデータであれば、CMSIS-DSP内のどの関数でも効果が見込めます。


cmsis_dsp/MatrixFunctions/arm_mat_mult_q31.c@1:fdd22bb7aa52, 2012-11-28 (annotated)

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

Who changed what in which revision?

UserRevisionLine numberNew contents of line
emilmont 1:fdd22bb7aa52 1 /* ----------------------------------------------------------------------
emilmont 1:fdd22bb7aa52 2 * Copyright (C) 2010 ARM Limited. All rights reserved.
emilmont 1:fdd22bb7aa52 3 *
emilmont 1:fdd22bb7aa52 4 * $Date: 15. February 2012
emilmont 1:fdd22bb7aa52 5 * $Revision: V1.1.0
emilmont 1:fdd22bb7aa52 6 *
emilmont 1:fdd22bb7aa52 7 * Project: CMSIS DSP Library
emilmont 1:fdd22bb7aa52 8 * Title: arm_mat_mult_q31.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 1:fdd22bb7aa52 10 * Description: Q31 matrix multiplication.
emilmont 1:fdd22bb7aa52 11 *
emilmont 1:fdd22bb7aa52 12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
emilmont 1:fdd22bb7aa52 13 *
emilmont 1:fdd22bb7aa52 14 * Version 1.1.0 2012/02/15
emilmont 1:fdd22bb7aa52 15 * Updated with more optimizations, bug fixes and minor API changes.
emilmont 1:fdd22bb7aa52 16 *
emilmont 1:fdd22bb7aa52 17 * Version 1.0.10 2011/7/15
emilmont 1:fdd22bb7aa52 18 * Big Endian support added and Merged M0 and M3/M4 Source code.
emilmont 1:fdd22bb7aa52 19 *
emilmont 1:fdd22bb7aa52 20 * Version 1.0.3 2010/11/29
emilmont 1:fdd22bb7aa52 21 * Re-organized the CMSIS folders and updated documentation.
emilmont 1:fdd22bb7aa52 22 *
emilmont 1:fdd22bb7aa52 23 * Version 1.0.2 2010/11/11
emilmont 1:fdd22bb7aa52 24 * Documentation updated.
emilmont 1:fdd22bb7aa52 25 *
emilmont 1:fdd22bb7aa52 26 * Version 1.0.1 2010/10/05
emilmont 1:fdd22bb7aa52 27 * Production release and review comments incorporated.
emilmont 1:fdd22bb7aa52 28 *
emilmont 1:fdd22bb7aa52 29 * Version 1.0.0 2010/09/20
emilmont 1:fdd22bb7aa52 30 * Production release and review comments incorporated.
emilmont 1:fdd22bb7aa52 31 *
emilmont 1:fdd22bb7aa52 32 * Version 0.0.5 2010/04/26
emilmont 1:fdd22bb7aa52 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 groupMatrix
emilmont 1:fdd22bb7aa52 43 */
emilmont 1:fdd22bb7aa52 44
emilmont 1:fdd22bb7aa52 45 /**
emilmont 1:fdd22bb7aa52 46 * @addtogroup MatrixMult
emilmont 1:fdd22bb7aa52 47 * @{
emilmont 1:fdd22bb7aa52 48 */
emilmont 1:fdd22bb7aa52 49
emilmont 1:fdd22bb7aa52 50 /**
emilmont 1:fdd22bb7aa52 51 * @brief Q31 matrix multiplication
emilmont 1:fdd22bb7aa52 52 * @param[in] *pSrcA points to the first input matrix structure
emilmont 1:fdd22bb7aa52 53 * @param[in] *pSrcB points to the second input matrix structure
emilmont 1:fdd22bb7aa52 54 * @param[out] *pDst points to output matrix structure
emilmont 1:fdd22bb7aa52 55 * @return The function returns either
emilmont 1:fdd22bb7aa52 56 * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
emilmont 1:fdd22bb7aa52 57 *
emilmont 1:fdd22bb7aa52 58 * @details
emilmont 1:fdd22bb7aa52 59 * <b>Scaling and Overflow Behavior:</b>
emilmont 1:fdd22bb7aa52 60 *
emilmont 1:fdd22bb7aa52 61 * \par
emilmont 1:fdd22bb7aa52 62 * The function is implemented using an internal 64-bit accumulator.
emilmont 1:fdd22bb7aa52 63 * The accumulator has a 2.62 format and maintains full precision of the intermediate
emilmont 1:fdd22bb7aa52 64 * multiplication results but provides only a single guard bit. There is no saturation
emilmont 1:fdd22bb7aa52 65 * on intermediate additions. Thus, if the accumulator overflows it wraps around and
emilmont 1:fdd22bb7aa52 66 * distorts the result. The input signals should be scaled down to avoid intermediate
emilmont 1:fdd22bb7aa52 67 * overflows. The input is thus scaled down by log2(numColsA) bits
emilmont 1:fdd22bb7aa52 68 * to avoid overflows, as a total of numColsA additions are performed internally.
emilmont 1:fdd22bb7aa52 69 * The 2.62 accumulator is right shifted by 31 bits and saturated to 1.31 format to yield the final result.
emilmont 1:fdd22bb7aa52 70 *
emilmont 1:fdd22bb7aa52 71 * \par
emilmont 1:fdd22bb7aa52 72 * See <code>arm_mat_mult_fast_q31()</code> for a faster but less precise implementation of this function for Cortex-M3 and Cortex-M4.
emilmont 1:fdd22bb7aa52 73 *
emilmont 1:fdd22bb7aa52 74 */
emilmont 1:fdd22bb7aa52 75
emilmont 1:fdd22bb7aa52 76 arm_status arm_mat_mult_q31(
emilmont 1:fdd22bb7aa52 77 const arm_matrix_instance_q31 * pSrcA,
emilmont 1:fdd22bb7aa52 78 const arm_matrix_instance_q31 * pSrcB,
emilmont 1:fdd22bb7aa52 79 arm_matrix_instance_q31 * pDst)
emilmont 1:fdd22bb7aa52 80 {
emilmont 1:fdd22bb7aa52 81 q31_t *pIn1 = pSrcA->pData; /* input data matrix pointer A */
emilmont 1:fdd22bb7aa52 82 q31_t *pIn2 = pSrcB->pData; /* input data matrix pointer B */
emilmont 1:fdd22bb7aa52 83 q31_t *pInA = pSrcA->pData; /* input data matrix pointer A */
emilmont 1:fdd22bb7aa52 84 q31_t *pOut = pDst->pData; /* output data matrix pointer */
emilmont 1:fdd22bb7aa52 85 q31_t *px; /* Temporary output data matrix pointer */
emilmont 1:fdd22bb7aa52 86 q63_t sum; /* Accumulator */
emilmont 1:fdd22bb7aa52 87 uint16_t numRowsA = pSrcA->numRows; /* number of rows of input matrix A */
emilmont 1:fdd22bb7aa52 88 uint16_t numColsB = pSrcB->numCols; /* number of columns of input matrix B */
emilmont 1:fdd22bb7aa52 89 uint16_t numColsA = pSrcA->numCols; /* number of columns of input matrix A */
emilmont 1:fdd22bb7aa52 90
emilmont 1:fdd22bb7aa52 91 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 92
emilmont 1:fdd22bb7aa52 93 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 94
emilmont 1:fdd22bb7aa52 95 uint16_t col, i = 0u, j, row = numRowsA, colCnt; /* loop counters */
emilmont 1:fdd22bb7aa52 96 arm_status status; /* status of matrix multiplication */
emilmont 1:fdd22bb7aa52 97 q31_t a0, a1, a2, a3, b0, b1, b2, b3;
emilmont 1:fdd22bb7aa52 98
emilmont 1:fdd22bb7aa52 99 #ifdef ARM_MATH_MATRIX_CHECK
emilmont 1:fdd22bb7aa52 100
emilmont 1:fdd22bb7aa52 101
emilmont 1:fdd22bb7aa52 102 /* Check for matrix mismatch condition */
emilmont 1:fdd22bb7aa52 103 if((pSrcA->numCols != pSrcB->numRows) ||
emilmont 1:fdd22bb7aa52 104 (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
emilmont 1:fdd22bb7aa52 105 {
emilmont 1:fdd22bb7aa52 106 /* Set status as ARM_MATH_SIZE_MISMATCH */
emilmont 1:fdd22bb7aa52 107 status = ARM_MATH_SIZE_MISMATCH;
emilmont 1:fdd22bb7aa52 108 }
emilmont 1:fdd22bb7aa52 109 else
emilmont 1:fdd22bb7aa52 110 #endif /* #ifdef ARM_MATH_MATRIX_CHECK */
emilmont 1:fdd22bb7aa52 111
emilmont 1:fdd22bb7aa52 112 {
emilmont 1:fdd22bb7aa52 113 /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
emilmont 1:fdd22bb7aa52 114 /* row loop */
emilmont 1:fdd22bb7aa52 115 do
emilmont 1:fdd22bb7aa52 116 {
emilmont 1:fdd22bb7aa52 117 /* Output pointer is set to starting address of the row being processed */
emilmont 1:fdd22bb7aa52 118 px = pOut + i;
emilmont 1:fdd22bb7aa52 119
emilmont 1:fdd22bb7aa52 120 /* For every row wise process, the column loop counter is to be initiated */
emilmont 1:fdd22bb7aa52 121 col = numColsB;
emilmont 1:fdd22bb7aa52 122
emilmont 1:fdd22bb7aa52 123 /* For every row wise process, the pIn2 pointer is set
emilmont 1:fdd22bb7aa52 124 ** to the starting address of the pSrcB data */
emilmont 1:fdd22bb7aa52 125 pIn2 = pSrcB->pData;
emilmont 1:fdd22bb7aa52 126
emilmont 1:fdd22bb7aa52 127 j = 0u;
emilmont 1:fdd22bb7aa52 128
emilmont 1:fdd22bb7aa52 129 /* column loop */
emilmont 1:fdd22bb7aa52 130 do
emilmont 1:fdd22bb7aa52 131 {
emilmont 1:fdd22bb7aa52 132 /* Set the variable sum, that acts as accumulator, to zero */
emilmont 1:fdd22bb7aa52 133 sum = 0;
emilmont 1:fdd22bb7aa52 134
emilmont 1:fdd22bb7aa52 135 /* Initiate the pointer pIn1 to point to the starting address of pInA */
emilmont 1:fdd22bb7aa52 136 pIn1 = pInA;
emilmont 1:fdd22bb7aa52 137
emilmont 1:fdd22bb7aa52 138 /* Apply loop unrolling and compute 4 MACs simultaneously. */
emilmont 1:fdd22bb7aa52 139 colCnt = numColsA >> 2;
emilmont 1:fdd22bb7aa52 140
emilmont 1:fdd22bb7aa52 141
emilmont 1:fdd22bb7aa52 142 /* matrix multiplication */
emilmont 1:fdd22bb7aa52 143 while(colCnt > 0u)
emilmont 1:fdd22bb7aa52 144 {
emilmont 1:fdd22bb7aa52 145 /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
emilmont 1:fdd22bb7aa52 146 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 147 b0 = *pIn2;
emilmont 1:fdd22bb7aa52 148 pIn2 += numColsB;
emilmont 1:fdd22bb7aa52 149
emilmont 1:fdd22bb7aa52 150 a0 = *pIn1++;
emilmont 1:fdd22bb7aa52 151 a1 = *pIn1++;
emilmont 1:fdd22bb7aa52 152
emilmont 1:fdd22bb7aa52 153 b1 = *pIn2;
emilmont 1:fdd22bb7aa52 154 pIn2 += numColsB;
emilmont 1:fdd22bb7aa52 155 b2 = *pIn2;
emilmont 1:fdd22bb7aa52 156 pIn2 += numColsB;
emilmont 1:fdd22bb7aa52 157
emilmont 1:fdd22bb7aa52 158 sum += (q63_t) a0 *b0;
emilmont 1:fdd22bb7aa52 159 sum += (q63_t) a1 *b1;
emilmont 1:fdd22bb7aa52 160
emilmont 1:fdd22bb7aa52 161 a2 = *pIn1++;
emilmont 1:fdd22bb7aa52 162 a3 = *pIn1++;
emilmont 1:fdd22bb7aa52 163
emilmont 1:fdd22bb7aa52 164 b3 = *pIn2;
emilmont 1:fdd22bb7aa52 165 pIn2 += numColsB;
emilmont 1:fdd22bb7aa52 166
emilmont 1:fdd22bb7aa52 167 sum += (q63_t) a2 *b2;
emilmont 1:fdd22bb7aa52 168 sum += (q63_t) a3 *b3;
emilmont 1:fdd22bb7aa52 169
emilmont 1:fdd22bb7aa52 170 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 171 colCnt--;
emilmont 1:fdd22bb7aa52 172 }
emilmont 1:fdd22bb7aa52 173
emilmont 1:fdd22bb7aa52 174 /* If the columns of pSrcA is not a multiple of 4, compute any remaining output samples here.
emilmont 1:fdd22bb7aa52 175 ** No loop unrolling is used. */
emilmont 1:fdd22bb7aa52 176 colCnt = numColsA % 0x4u;
emilmont 1:fdd22bb7aa52 177
emilmont 1:fdd22bb7aa52 178 while(colCnt > 0u)
emilmont 1:fdd22bb7aa52 179 {
emilmont 1:fdd22bb7aa52 180 /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
emilmont 1:fdd22bb7aa52 181 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 182 sum += (q63_t) * pIn1++ * *pIn2;
emilmont 1:fdd22bb7aa52 183 pIn2 += numColsB;
emilmont 1:fdd22bb7aa52 184
emilmont 1:fdd22bb7aa52 185 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 186 colCnt--;
emilmont 1:fdd22bb7aa52 187 }
emilmont 1:fdd22bb7aa52 188
emilmont 1:fdd22bb7aa52 189 /* Convert the result from 2.62 to 1.31 format and store in destination buffer */
emilmont 1:fdd22bb7aa52 190 *px++ = (q31_t) (sum >> 31);
emilmont 1:fdd22bb7aa52 191
emilmont 1:fdd22bb7aa52 192 /* Update the pointer pIn2 to point to the starting address of the next column */
emilmont 1:fdd22bb7aa52 193 j++;
emilmont 1:fdd22bb7aa52 194 pIn2 = (pSrcB->pData) + j;
emilmont 1:fdd22bb7aa52 195
emilmont 1:fdd22bb7aa52 196 /* Decrement the column loop counter */
emilmont 1:fdd22bb7aa52 197 col--;
emilmont 1:fdd22bb7aa52 198
emilmont 1:fdd22bb7aa52 199 } while(col > 0u);
emilmont 1:fdd22bb7aa52 200
emilmont 1:fdd22bb7aa52 201 #else
emilmont 1:fdd22bb7aa52 202
emilmont 1:fdd22bb7aa52 203 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 204
emilmont 1:fdd22bb7aa52 205 q31_t *pInB = pSrcB->pData; /* input data matrix pointer B */
emilmont 1:fdd22bb7aa52 206 uint16_t col, i = 0u, row = numRowsA, colCnt; /* loop counters */
emilmont 1:fdd22bb7aa52 207 arm_status status; /* status of matrix multiplication */
emilmont 1:fdd22bb7aa52 208
emilmont 1:fdd22bb7aa52 209
emilmont 1:fdd22bb7aa52 210 #ifdef ARM_MATH_MATRIX_CHECK
emilmont 1:fdd22bb7aa52 211
emilmont 1:fdd22bb7aa52 212 /* Check for matrix mismatch condition */
emilmont 1:fdd22bb7aa52 213 if((pSrcA->numCols != pSrcB->numRows) ||
emilmont 1:fdd22bb7aa52 214 (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
emilmont 1:fdd22bb7aa52 215 {
emilmont 1:fdd22bb7aa52 216 /* Set status as ARM_MATH_SIZE_MISMATCH */
emilmont 1:fdd22bb7aa52 217 status = ARM_MATH_SIZE_MISMATCH;
emilmont 1:fdd22bb7aa52 218 }
emilmont 1:fdd22bb7aa52 219 else
emilmont 1:fdd22bb7aa52 220 #endif /* #ifdef ARM_MATH_MATRIX_CHECK */
emilmont 1:fdd22bb7aa52 221
emilmont 1:fdd22bb7aa52 222 {
emilmont 1:fdd22bb7aa52 223 /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
emilmont 1:fdd22bb7aa52 224 /* row loop */
emilmont 1:fdd22bb7aa52 225 do
emilmont 1:fdd22bb7aa52 226 {
emilmont 1:fdd22bb7aa52 227 /* Output pointer is set to starting address of the row being processed */
emilmont 1:fdd22bb7aa52 228 px = pOut + i;
emilmont 1:fdd22bb7aa52 229
emilmont 1:fdd22bb7aa52 230 /* For every row wise process, the column loop counter is to be initiated */
emilmont 1:fdd22bb7aa52 231 col = numColsB;
emilmont 1:fdd22bb7aa52 232
emilmont 1:fdd22bb7aa52 233 /* For every row wise process, the pIn2 pointer is set
emilmont 1:fdd22bb7aa52 234 ** to the starting address of the pSrcB data */
emilmont 1:fdd22bb7aa52 235 pIn2 = pSrcB->pData;
emilmont 1:fdd22bb7aa52 236
emilmont 1:fdd22bb7aa52 237 /* column loop */
emilmont 1:fdd22bb7aa52 238 do
emilmont 1:fdd22bb7aa52 239 {
emilmont 1:fdd22bb7aa52 240 /* Set the variable sum, that acts as accumulator, to zero */
emilmont 1:fdd22bb7aa52 241 sum = 0;
emilmont 1:fdd22bb7aa52 242
emilmont 1:fdd22bb7aa52 243 /* Initiate the pointer pIn1 to point to the starting address of pInA */
emilmont 1:fdd22bb7aa52 244 pIn1 = pInA;
emilmont 1:fdd22bb7aa52 245
emilmont 1:fdd22bb7aa52 246 /* Matrix A columns number of MAC operations are to be performed */
emilmont 1:fdd22bb7aa52 247 colCnt = numColsA;
emilmont 1:fdd22bb7aa52 248
emilmont 1:fdd22bb7aa52 249 /* matrix multiplication */
emilmont 1:fdd22bb7aa52 250 while(colCnt > 0u)
emilmont 1:fdd22bb7aa52 251 {
emilmont 1:fdd22bb7aa52 252 /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
emilmont 1:fdd22bb7aa52 253 /* Perform the multiply-accumulates */
emilmont 1:fdd22bb7aa52 254 sum += (q63_t) * pIn1++ * *pIn2;
emilmont 1:fdd22bb7aa52 255 pIn2 += numColsB;
emilmont 1:fdd22bb7aa52 256
emilmont 1:fdd22bb7aa52 257 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 258 colCnt--;
emilmont 1:fdd22bb7aa52 259 }
emilmont 1:fdd22bb7aa52 260
emilmont 1:fdd22bb7aa52 261 /* Convert the result from 2.62 to 1.31 format and store in destination buffer */
emilmont 1:fdd22bb7aa52 262 *px++ = (q31_t) (sum >> 31);
emilmont 1:fdd22bb7aa52 263
emilmont 1:fdd22bb7aa52 264 /* Decrement the column loop counter */
emilmont 1:fdd22bb7aa52 265 col--;
emilmont 1:fdd22bb7aa52 266
emilmont 1:fdd22bb7aa52 267 /* Update the pointer pIn2 to point to the starting address of the next column */
emilmont 1:fdd22bb7aa52 268 pIn2 = pInB + (numColsB - col);
emilmont 1:fdd22bb7aa52 269
emilmont 1:fdd22bb7aa52 270 } while(col > 0u);
emilmont 1:fdd22bb7aa52 271
emilmont 1:fdd22bb7aa52 272 #endif
emilmont 1:fdd22bb7aa52 273
emilmont 1:fdd22bb7aa52 274 /* Update the pointer pInA to point to the starting address of the next row */
emilmont 1:fdd22bb7aa52 275 i = i + numColsB;
emilmont 1:fdd22bb7aa52 276 pInA = pInA + numColsA;
emilmont 1:fdd22bb7aa52 277
emilmont 1:fdd22bb7aa52 278 /* Decrement the row loop counter */
emilmont 1:fdd22bb7aa52 279 row--;
emilmont 1:fdd22bb7aa52 280
emilmont 1:fdd22bb7aa52 281 } while(row > 0u);
emilmont 1:fdd22bb7aa52 282
emilmont 1:fdd22bb7aa52 283 /* set status as ARM_MATH_SUCCESS */
emilmont 1:fdd22bb7aa52 284 status = ARM_MATH_SUCCESS;
emilmont 1:fdd22bb7aa52 285 }
emilmont 1:fdd22bb7aa52 286 /* Return to application */
emilmont 1:fdd22bb7aa52 287 return (status);
emilmont 1:fdd22bb7aa52 288 }
emilmont 1:fdd22bb7aa52 289
emilmont 1:fdd22bb7aa52 290 /**
emilmont 1:fdd22bb7aa52 291 * @} end of MatrixMult group
emilmont 1:fdd22bb7aa52 292 */