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。


Committer:
emilmont
Date:
Thu May 30 17:10:11 2013 +0100
Revision:
2:da51fb522205
Parent:
1:fdd22bb7aa52
Child:
3:7a284390b0ce
Keep "cmsis-dsp" module in synch with its source

Who changed what in which revision?

UserRevisionLine numberNew contents of line
emilmont 1:fdd22bb7aa52 1 /* ----------------------------------------------------------------------
emilmont 1:fdd22bb7aa52 2 * Copyright (C) 2010 ARM Limited. All rights reserved.
emilmont 1:fdd22bb7aa52 3 *
emilmont 1:fdd22bb7aa52 4 * $Date: 15. February 2012
emilmont 2:da51fb522205 5 * $Revision: V1.1.0
emilmont 1:fdd22bb7aa52 6 *
emilmont 2:da51fb522205 7 * Project: CMSIS DSP Library
emilmont 2:da51fb522205 8 * Title: arm_max_q31.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 2:da51fb522205 10 * Description: Maximum value of a Q31 vector.
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
emilmont 1:fdd22bb7aa52 33 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 34
emilmont 1:fdd22bb7aa52 35 /**
emilmont 1:fdd22bb7aa52 36 * @ingroup groupStats
emilmont 1:fdd22bb7aa52 37 */
emilmont 1:fdd22bb7aa52 38
emilmont 1:fdd22bb7aa52 39 /**
emilmont 1:fdd22bb7aa52 40 * @addtogroup Max
emilmont 1:fdd22bb7aa52 41 * @{
emilmont 1:fdd22bb7aa52 42 */
emilmont 1:fdd22bb7aa52 43
emilmont 1:fdd22bb7aa52 44
emilmont 1:fdd22bb7aa52 45 /**
emilmont 1:fdd22bb7aa52 46 * @brief Maximum value of a Q31 vector.
emilmont 1:fdd22bb7aa52 47 * @param[in] *pSrc points to the input vector
emilmont 1:fdd22bb7aa52 48 * @param[in] blockSize length of the input vector
emilmont 1:fdd22bb7aa52 49 * @param[out] *pResult maximum value returned here
emilmont 1:fdd22bb7aa52 50 * @param[out] *pIndex index of maximum value returned here
emilmont 1:fdd22bb7aa52 51 * @return none.
emilmont 1:fdd22bb7aa52 52 */
emilmont 1:fdd22bb7aa52 53
emilmont 1:fdd22bb7aa52 54 void arm_max_q31(
emilmont 1:fdd22bb7aa52 55 q31_t * pSrc,
emilmont 1:fdd22bb7aa52 56 uint32_t blockSize,
emilmont 1:fdd22bb7aa52 57 q31_t * pResult,
emilmont 1:fdd22bb7aa52 58 uint32_t * pIndex)
emilmont 1:fdd22bb7aa52 59 {
emilmont 1:fdd22bb7aa52 60 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 61
emilmont 1:fdd22bb7aa52 62 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 63 q31_t maxVal1, maxVal2, out; /* Temporary variables to store the output value. */
emilmont 1:fdd22bb7aa52 64 uint32_t blkCnt, outIndex, count; /* loop counter */
emilmont 1:fdd22bb7aa52 65
emilmont 1:fdd22bb7aa52 66 /* Initialise the count value. */
emilmont 1:fdd22bb7aa52 67 count = 0u;
emilmont 1:fdd22bb7aa52 68 /* Initialise the index value to zero. */
emilmont 1:fdd22bb7aa52 69 outIndex = 0u;
emilmont 1:fdd22bb7aa52 70 /* Load first input value that act as reference value for comparision */
emilmont 1:fdd22bb7aa52 71 out = *pSrc++;
emilmont 1:fdd22bb7aa52 72
emilmont 1:fdd22bb7aa52 73 /* Loop unrolling */
emilmont 1:fdd22bb7aa52 74 blkCnt = (blockSize - 1u) >> 2u;
emilmont 1:fdd22bb7aa52 75
emilmont 1:fdd22bb7aa52 76 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 77 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 78 {
emilmont 1:fdd22bb7aa52 79 /* Initialize maxVal to the next consecutive values one by one */
emilmont 1:fdd22bb7aa52 80 maxVal1 = *pSrc++;
emilmont 1:fdd22bb7aa52 81
emilmont 1:fdd22bb7aa52 82 maxVal2 = *pSrc++;
emilmont 1:fdd22bb7aa52 83
emilmont 1:fdd22bb7aa52 84 /* compare for the maximum value */
emilmont 1:fdd22bb7aa52 85 if(out < maxVal1)
emilmont 1:fdd22bb7aa52 86 {
emilmont 1:fdd22bb7aa52 87 /* Update the maximum value and its index */
emilmont 1:fdd22bb7aa52 88 out = maxVal1;
emilmont 1:fdd22bb7aa52 89 outIndex = count + 1u;
emilmont 1:fdd22bb7aa52 90 }
emilmont 1:fdd22bb7aa52 91
emilmont 1:fdd22bb7aa52 92 maxVal1 = *pSrc++;
emilmont 1:fdd22bb7aa52 93
emilmont 1:fdd22bb7aa52 94 /* compare for the maximum value */
emilmont 1:fdd22bb7aa52 95 if(out < maxVal2)
emilmont 1:fdd22bb7aa52 96 {
emilmont 1:fdd22bb7aa52 97 /* Update the maximum value and its index */
emilmont 1:fdd22bb7aa52 98 out = maxVal2;
emilmont 1:fdd22bb7aa52 99 outIndex = count + 2u;
emilmont 1:fdd22bb7aa52 100 }
emilmont 1:fdd22bb7aa52 101
emilmont 1:fdd22bb7aa52 102 maxVal2 = *pSrc++;
emilmont 1:fdd22bb7aa52 103
emilmont 1:fdd22bb7aa52 104 /* compare for the maximum value */
emilmont 1:fdd22bb7aa52 105 if(out < maxVal1)
emilmont 1:fdd22bb7aa52 106 {
emilmont 1:fdd22bb7aa52 107 /* Update the maximum value and its index */
emilmont 1:fdd22bb7aa52 108 out = maxVal1;
emilmont 1:fdd22bb7aa52 109 outIndex = count + 3u;
emilmont 1:fdd22bb7aa52 110 }
emilmont 1:fdd22bb7aa52 111
emilmont 1:fdd22bb7aa52 112 /* compare for the maximum value */
emilmont 1:fdd22bb7aa52 113 if(out < maxVal2)
emilmont 1:fdd22bb7aa52 114 {
emilmont 1:fdd22bb7aa52 115 /* Update the maximum value and its index */
emilmont 1:fdd22bb7aa52 116 out = maxVal2;
emilmont 1:fdd22bb7aa52 117 outIndex = count + 4u;
emilmont 1:fdd22bb7aa52 118 }
emilmont 1:fdd22bb7aa52 119
emilmont 1:fdd22bb7aa52 120 count += 4u;
emilmont 1:fdd22bb7aa52 121
emilmont 1:fdd22bb7aa52 122 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 123 blkCnt--;
emilmont 1:fdd22bb7aa52 124 }
emilmont 1:fdd22bb7aa52 125
emilmont 1:fdd22bb7aa52 126 /* if (blockSize - 1u) is not multiple of 4 */
emilmont 1:fdd22bb7aa52 127 blkCnt = (blockSize - 1u) % 4u;
emilmont 1:fdd22bb7aa52 128
emilmont 1:fdd22bb7aa52 129 #else
emilmont 1:fdd22bb7aa52 130
emilmont 1:fdd22bb7aa52 131 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 132 q31_t maxVal1, out; /* Temporary variables to store the output value. */
emilmont 1:fdd22bb7aa52 133 uint32_t blkCnt, outIndex; /* loop counter */
emilmont 1:fdd22bb7aa52 134
emilmont 1:fdd22bb7aa52 135 /* Initialise the index value to zero. */
emilmont 1:fdd22bb7aa52 136 outIndex = 0u;
emilmont 1:fdd22bb7aa52 137 /* Load first input value that act as reference value for comparision */
emilmont 1:fdd22bb7aa52 138 out = *pSrc++;
emilmont 1:fdd22bb7aa52 139
emilmont 1:fdd22bb7aa52 140 blkCnt = (blockSize - 1u);
emilmont 1:fdd22bb7aa52 141
emilmont 1:fdd22bb7aa52 142 #endif /* #ifndef ARM_MATH_CM0 */
emilmont 1:fdd22bb7aa52 143
emilmont 1:fdd22bb7aa52 144 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 145 {
emilmont 1:fdd22bb7aa52 146 /* Initialize maxVal to the next consecutive values one by one */
emilmont 1:fdd22bb7aa52 147 maxVal1 = *pSrc++;
emilmont 1:fdd22bb7aa52 148
emilmont 1:fdd22bb7aa52 149 /* compare for the maximum value */
emilmont 1:fdd22bb7aa52 150 if(out < maxVal1)
emilmont 1:fdd22bb7aa52 151 {
emilmont 1:fdd22bb7aa52 152 /* Update the maximum value and it's index */
emilmont 1:fdd22bb7aa52 153 out = maxVal1;
emilmont 1:fdd22bb7aa52 154 outIndex = blockSize - blkCnt;
emilmont 1:fdd22bb7aa52 155 }
emilmont 1:fdd22bb7aa52 156
emilmont 1:fdd22bb7aa52 157 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 158 blkCnt--;
emilmont 1:fdd22bb7aa52 159
emilmont 1:fdd22bb7aa52 160 }
emilmont 1:fdd22bb7aa52 161
emilmont 1:fdd22bb7aa52 162 /* Store the maximum value and its index into destination pointers */
emilmont 1:fdd22bb7aa52 163 *pResult = out;
emilmont 1:fdd22bb7aa52 164 *pIndex = outIndex;
emilmont 1:fdd22bb7aa52 165 }
emilmont 1:fdd22bb7aa52 166
emilmont 1:fdd22bb7aa52 167 /**
emilmont 1:fdd22bb7aa52 168 * @} end of Max group
emilmont 1:fdd22bb7aa52 169 */