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/BasicMathFunctions/arm_offset_f32.c@2:da51fb522205, 2013-05-30 (annotated)

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

Who changed what in which revision?

UserRevisionLine numberNew contents of line
emilmont 1:fdd22bb7aa52 1 /* ----------------------------------------------------------------------
emilmont 1:fdd22bb7aa52 2 * Copyright (C) 2010 ARM Limited. All rights reserved.
emilmont 1:fdd22bb7aa52 3 *
emilmont 1:fdd22bb7aa52 4 * $Date: 15. February 2012
emilmont 2:da51fb522205 5 * $Revision: V1.1.0
emilmont 1:fdd22bb7aa52 6 *
emilmont 2:da51fb522205 7 * Project: CMSIS DSP Library
emilmont 2:da51fb522205 8 * Title: arm_offset_f32.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 2:da51fb522205 10 * Description: Floating-point vector offset.
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 0.0.7 2010/06/10
emilmont 1:fdd22bb7aa52 33 * Misra-C changes done
emilmont 1:fdd22bb7aa52 34 * ---------------------------------------------------------------------------- */
emilmont 1:fdd22bb7aa52 35 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 36
emilmont 1:fdd22bb7aa52 37 /**
emilmont 1:fdd22bb7aa52 38 * @ingroup groupMath
emilmont 1:fdd22bb7aa52 39 */
emilmont 1:fdd22bb7aa52 40
emilmont 1:fdd22bb7aa52 41 /**
emilmont 1:fdd22bb7aa52 42 * @defgroup offset Vector Offset
emilmont 1:fdd22bb7aa52 43 *
emilmont 1:fdd22bb7aa52 44 * Adds a constant offset to each element of a vector.
emilmont 1:fdd22bb7aa52 45 *
emilmont 1:fdd22bb7aa52 46 * <pre>
emilmont 1:fdd22bb7aa52 47 * pDst[n] = pSrc[n] + offset, 0 <= n < blockSize.
emilmont 1:fdd22bb7aa52 48 * </pre>
emilmont 1:fdd22bb7aa52 49 *
emilmont 1:fdd22bb7aa52 50 * There are separate functions for floating-point, Q7, Q15, and Q31 data types.
emilmont 1:fdd22bb7aa52 51 */
emilmont 1:fdd22bb7aa52 52
emilmont 1:fdd22bb7aa52 53 /**
emilmont 1:fdd22bb7aa52 54 * @addtogroup offset
emilmont 1:fdd22bb7aa52 55 * @{
emilmont 1:fdd22bb7aa52 56 */
emilmont 1:fdd22bb7aa52 57
emilmont 1:fdd22bb7aa52 58 /**
emilmont 1:fdd22bb7aa52 59 * @brief Adds a constant offset to a floating-point vector.
emilmont 1:fdd22bb7aa52 60 * @param[in] *pSrc points to the input vector
emilmont 1:fdd22bb7aa52 61 * @param[in] offset is the offset to be added
emilmont 1:fdd22bb7aa52 62 * @param[out] *pDst points to the output vector
emilmont 1:fdd22bb7aa52 63 * @param[in] blockSize number of samples in the vector
emilmont 1:fdd22bb7aa52 64 * @return none.
emilmont 1:fdd22bb7aa52 65 */
emilmont 1:fdd22bb7aa52 66
emilmont 1:fdd22bb7aa52 67
emilmont 1:fdd22bb7aa52 68 void arm_offset_f32(
emilmont 1:fdd22bb7aa52 69 float32_t * pSrc,
emilmont 1:fdd22bb7aa52 70 float32_t offset,
emilmont 1:fdd22bb7aa52 71 float32_t * pDst,
emilmont 1:fdd22bb7aa52 72 uint32_t blockSize)
emilmont 1:fdd22bb7aa52 73 {
emilmont 1:fdd22bb7aa52 74 uint32_t blkCnt; /* loop counter */
emilmont 1:fdd22bb7aa52 75
emilmont 1:fdd22bb7aa52 76 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 77
emilmont 1:fdd22bb7aa52 78 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 79 float32_t in1, in2, in3, in4;
emilmont 1:fdd22bb7aa52 80
emilmont 1:fdd22bb7aa52 81 /*loop Unrolling */
emilmont 1:fdd22bb7aa52 82 blkCnt = blockSize >> 2u;
emilmont 1:fdd22bb7aa52 83
emilmont 1:fdd22bb7aa52 84 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
emilmont 1:fdd22bb7aa52 85 ** a second loop below computes the remaining 1 to 3 samples. */
emilmont 1:fdd22bb7aa52 86 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 87 {
emilmont 1:fdd22bb7aa52 88 /* C = A + offset */
emilmont 1:fdd22bb7aa52 89 /* Add offset and then store the results in the destination buffer. */
emilmont 1:fdd22bb7aa52 90 /* read samples from source */
emilmont 1:fdd22bb7aa52 91 in1 = *pSrc;
emilmont 1:fdd22bb7aa52 92 in2 = *(pSrc + 1);
emilmont 1:fdd22bb7aa52 93
emilmont 1:fdd22bb7aa52 94 /* add offset to input */
emilmont 1:fdd22bb7aa52 95 in1 = in1 + offset;
emilmont 1:fdd22bb7aa52 96
emilmont 1:fdd22bb7aa52 97 /* read samples from source */
emilmont 1:fdd22bb7aa52 98 in3 = *(pSrc + 2);
emilmont 1:fdd22bb7aa52 99
emilmont 1:fdd22bb7aa52 100 /* add offset to input */
emilmont 1:fdd22bb7aa52 101 in2 = in2 + offset;
emilmont 1:fdd22bb7aa52 102
emilmont 1:fdd22bb7aa52 103 /* read samples from source */
emilmont 1:fdd22bb7aa52 104 in4 = *(pSrc + 3);
emilmont 1:fdd22bb7aa52 105
emilmont 1:fdd22bb7aa52 106 /* add offset to input */
emilmont 1:fdd22bb7aa52 107 in3 = in3 + offset;
emilmont 1:fdd22bb7aa52 108
emilmont 1:fdd22bb7aa52 109 /* store result to destination */
emilmont 1:fdd22bb7aa52 110 *pDst = in1;
emilmont 1:fdd22bb7aa52 111
emilmont 1:fdd22bb7aa52 112 /* add offset to input */
emilmont 1:fdd22bb7aa52 113 in4 = in4 + offset;
emilmont 1:fdd22bb7aa52 114
emilmont 1:fdd22bb7aa52 115 /* store result to destination */
emilmont 1:fdd22bb7aa52 116 *(pDst + 1) = in2;
emilmont 1:fdd22bb7aa52 117
emilmont 1:fdd22bb7aa52 118 /* store result to destination */
emilmont 1:fdd22bb7aa52 119 *(pDst + 2) = in3;
emilmont 1:fdd22bb7aa52 120
emilmont 1:fdd22bb7aa52 121 /* store result to destination */
emilmont 1:fdd22bb7aa52 122 *(pDst + 3) = in4;
emilmont 1:fdd22bb7aa52 123
emilmont 1:fdd22bb7aa52 124 /* update pointers to process next samples */
emilmont 1:fdd22bb7aa52 125 pSrc += 4u;
emilmont 1:fdd22bb7aa52 126 pDst += 4u;
emilmont 1:fdd22bb7aa52 127
emilmont 1:fdd22bb7aa52 128 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 129 blkCnt--;
emilmont 1:fdd22bb7aa52 130 }
emilmont 1:fdd22bb7aa52 131
emilmont 1:fdd22bb7aa52 132 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
emilmont 1:fdd22bb7aa52 133 ** No loop unrolling is used. */
emilmont 1:fdd22bb7aa52 134 blkCnt = blockSize % 0x4u;
emilmont 1:fdd22bb7aa52 135
emilmont 1:fdd22bb7aa52 136 #else
emilmont 1:fdd22bb7aa52 137
emilmont 1:fdd22bb7aa52 138 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 139
emilmont 1:fdd22bb7aa52 140 /* Initialize blkCnt with number of samples */
emilmont 1:fdd22bb7aa52 141 blkCnt = blockSize;
emilmont 1:fdd22bb7aa52 142
emilmont 1:fdd22bb7aa52 143 #endif /* #ifndef ARM_MATH_CM0 */
emilmont 1:fdd22bb7aa52 144
emilmont 1:fdd22bb7aa52 145 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 146 {
emilmont 1:fdd22bb7aa52 147 /* C = A + offset */
emilmont 1:fdd22bb7aa52 148 /* Add offset and then store the result in the destination buffer. */
emilmont 1:fdd22bb7aa52 149 *pDst++ = (*pSrc++) + offset;
emilmont 1:fdd22bb7aa52 150
emilmont 1:fdd22bb7aa52 151 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 152 blkCnt--;
emilmont 1:fdd22bb7aa52 153 }
emilmont 1:fdd22bb7aa52 154 }
emilmont 1:fdd22bb7aa52 155
emilmont 1:fdd22bb7aa52 156 /**
emilmont 1:fdd22bb7aa52 157 * @} end of offset group
emilmont 1:fdd22bb7aa52 158 */