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/ComplexMathFunctions/arm_cmplx_conj_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_cmplx_conj_f32.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 2:da51fb522205 10 * Description: Floating-point complex conjugate.
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 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 33
emilmont 1:fdd22bb7aa52 34 /**
emilmont 1:fdd22bb7aa52 35 * @ingroup groupCmplxMath
emilmont 1:fdd22bb7aa52 36 */
emilmont 1:fdd22bb7aa52 37
emilmont 1:fdd22bb7aa52 38 /**
emilmont 1:fdd22bb7aa52 39 * @defgroup cmplx_conj Complex Conjugate
emilmont 1:fdd22bb7aa52 40 *
emilmont 1:fdd22bb7aa52 41 * Conjugates the elements of a complex data vector.
emilmont 1:fdd22bb7aa52 42 *
emilmont 1:fdd22bb7aa52 43 * The <code>pSrc</code> points to the source data and
emilmont 1:fdd22bb7aa52 44 * <code>pDst</code> points to the where the result should be written.
emilmont 1:fdd22bb7aa52 45 * <code>numSamples</code> specifies the number of complex samples
emilmont 1:fdd22bb7aa52 46 * and the data in each array is stored in an interleaved fashion
emilmont 1:fdd22bb7aa52 47 * (real, imag, real, imag, ...).
emilmont 1:fdd22bb7aa52 48 * Each array has a total of <code>2*numSamples</code> values.
emilmont 1:fdd22bb7aa52 49 * The underlying algorithm is used:
emilmont 1:fdd22bb7aa52 50 *
emilmont 1:fdd22bb7aa52 51 * <pre>
emilmont 1:fdd22bb7aa52 52 * for(n=0; n<numSamples; n++) {
emilmont 1:fdd22bb7aa52 53 * pDst[(2*n)+0)] = pSrc[(2*n)+0]; // real part
emilmont 1:fdd22bb7aa52 54 * pDst[(2*n)+1)] = -pSrc[(2*n)+1]; // imag part
emilmont 1:fdd22bb7aa52 55 * }
emilmont 1:fdd22bb7aa52 56 * </pre>
emilmont 1:fdd22bb7aa52 57 *
emilmont 1:fdd22bb7aa52 58 * There are separate functions for floating-point, Q15, and Q31 data types.
emilmont 1:fdd22bb7aa52 59 */
emilmont 1:fdd22bb7aa52 60
emilmont 1:fdd22bb7aa52 61 /**
emilmont 1:fdd22bb7aa52 62 * @addtogroup cmplx_conj
emilmont 1:fdd22bb7aa52 63 * @{
emilmont 1:fdd22bb7aa52 64 */
emilmont 1:fdd22bb7aa52 65
emilmont 1:fdd22bb7aa52 66 /**
emilmont 1:fdd22bb7aa52 67 * @brief Floating-point complex conjugate.
emilmont 1:fdd22bb7aa52 68 * @param *pSrc points to the input vector
emilmont 1:fdd22bb7aa52 69 * @param *pDst points to the output vector
emilmont 1:fdd22bb7aa52 70 * @param numSamples number of complex samples in each vector
emilmont 1:fdd22bb7aa52 71 * @return none.
emilmont 1:fdd22bb7aa52 72 */
emilmont 1:fdd22bb7aa52 73
emilmont 1:fdd22bb7aa52 74 void arm_cmplx_conj_f32(
emilmont 1:fdd22bb7aa52 75 float32_t * pSrc,
emilmont 1:fdd22bb7aa52 76 float32_t * pDst,
emilmont 1:fdd22bb7aa52 77 uint32_t numSamples)
emilmont 1:fdd22bb7aa52 78 {
emilmont 1:fdd22bb7aa52 79 uint32_t blkCnt; /* loop counter */
emilmont 1:fdd22bb7aa52 80
emilmont 1:fdd22bb7aa52 81 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 82
emilmont 1:fdd22bb7aa52 83 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 84 float32_t inR1, inR2, inR3, inR4;
emilmont 1:fdd22bb7aa52 85 float32_t inI1, inI2, inI3, inI4;
emilmont 1:fdd22bb7aa52 86
emilmont 1:fdd22bb7aa52 87 /*loop Unrolling */
emilmont 1:fdd22bb7aa52 88 blkCnt = numSamples >> 2u;
emilmont 1:fdd22bb7aa52 89
emilmont 1:fdd22bb7aa52 90 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
emilmont 1:fdd22bb7aa52 91 ** a second loop below computes the remaining 1 to 3 samples. */
emilmont 1:fdd22bb7aa52 92 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 93 {
emilmont 1:fdd22bb7aa52 94 /* C[0]+jC[1] = A[0]+ j (-1) A[1] */
emilmont 1:fdd22bb7aa52 95 /* Calculate Complex Conjugate and then store the results in the destination buffer. */
emilmont 1:fdd22bb7aa52 96 /* read real input samples */
emilmont 1:fdd22bb7aa52 97 inR1 = pSrc[0];
emilmont 1:fdd22bb7aa52 98 /* store real samples to destination */
emilmont 1:fdd22bb7aa52 99 pDst[0] = inR1;
emilmont 1:fdd22bb7aa52 100 inR2 = pSrc[2];
emilmont 1:fdd22bb7aa52 101 pDst[2] = inR2;
emilmont 1:fdd22bb7aa52 102 inR3 = pSrc[4];
emilmont 1:fdd22bb7aa52 103 pDst[4] = inR3;
emilmont 1:fdd22bb7aa52 104 inR4 = pSrc[6];
emilmont 1:fdd22bb7aa52 105 pDst[6] = inR4;
emilmont 1:fdd22bb7aa52 106
emilmont 1:fdd22bb7aa52 107 /* read imaginary input samples */
emilmont 1:fdd22bb7aa52 108 inI1 = pSrc[1];
emilmont 1:fdd22bb7aa52 109 inI2 = pSrc[3];
emilmont 1:fdd22bb7aa52 110
emilmont 1:fdd22bb7aa52 111 /* conjugate input */
emilmont 1:fdd22bb7aa52 112 inI1 = -inI1;
emilmont 1:fdd22bb7aa52 113
emilmont 1:fdd22bb7aa52 114 /* read imaginary input samples */
emilmont 1:fdd22bb7aa52 115 inI3 = pSrc[5];
emilmont 1:fdd22bb7aa52 116
emilmont 1:fdd22bb7aa52 117 /* conjugate input */
emilmont 1:fdd22bb7aa52 118 inI2 = -inI2;
emilmont 1:fdd22bb7aa52 119
emilmont 1:fdd22bb7aa52 120 /* read imaginary input samples */
emilmont 1:fdd22bb7aa52 121 inI4 = pSrc[7];
emilmont 1:fdd22bb7aa52 122
emilmont 1:fdd22bb7aa52 123 /* conjugate input */
emilmont 1:fdd22bb7aa52 124 inI3 = -inI3;
emilmont 1:fdd22bb7aa52 125
emilmont 1:fdd22bb7aa52 126 /* store imaginary samples to destination */
emilmont 1:fdd22bb7aa52 127 pDst[1] = inI1;
emilmont 1:fdd22bb7aa52 128 pDst[3] = inI2;
emilmont 1:fdd22bb7aa52 129
emilmont 1:fdd22bb7aa52 130 /* conjugate input */
emilmont 1:fdd22bb7aa52 131 inI4 = -inI4;
emilmont 1:fdd22bb7aa52 132
emilmont 1:fdd22bb7aa52 133 /* store imaginary samples to destination */
emilmont 1:fdd22bb7aa52 134 pDst[5] = inI3;
emilmont 1:fdd22bb7aa52 135
emilmont 1:fdd22bb7aa52 136 /* increment source pointer by 8 to process next sampels */
emilmont 1:fdd22bb7aa52 137 pSrc += 8u;
emilmont 1:fdd22bb7aa52 138
emilmont 1:fdd22bb7aa52 139 /* store imaginary sample to destination */
emilmont 1:fdd22bb7aa52 140 pDst[7] = inI4;
emilmont 1:fdd22bb7aa52 141
emilmont 1:fdd22bb7aa52 142 /* increment destination pointer by 8 to store next samples */
emilmont 1:fdd22bb7aa52 143 pDst += 8u;
emilmont 1:fdd22bb7aa52 144
emilmont 1:fdd22bb7aa52 145 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 146 blkCnt--;
emilmont 1:fdd22bb7aa52 147 }
emilmont 1:fdd22bb7aa52 148
emilmont 1:fdd22bb7aa52 149 /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
emilmont 1:fdd22bb7aa52 150 ** No loop unrolling is used. */
emilmont 1:fdd22bb7aa52 151 blkCnt = numSamples % 0x4u;
emilmont 1:fdd22bb7aa52 152
emilmont 1:fdd22bb7aa52 153 #else
emilmont 1:fdd22bb7aa52 154
emilmont 1:fdd22bb7aa52 155 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 156 blkCnt = numSamples;
emilmont 1:fdd22bb7aa52 157
emilmont 1:fdd22bb7aa52 158 #endif /* #ifndef ARM_MATH_CM0 */
emilmont 1:fdd22bb7aa52 159
emilmont 1:fdd22bb7aa52 160 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 161 {
emilmont 1:fdd22bb7aa52 162 /* realOut + j (imagOut) = realIn + j (-1) imagIn */
emilmont 1:fdd22bb7aa52 163 /* Calculate Complex Conjugate and then store the results in the destination buffer. */
emilmont 1:fdd22bb7aa52 164 *pDst++ = *pSrc++;
emilmont 1:fdd22bb7aa52 165 *pDst++ = -*pSrc++;
emilmont 1:fdd22bb7aa52 166
emilmont 1:fdd22bb7aa52 167 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 168 blkCnt--;
emilmont 1:fdd22bb7aa52 169 }
emilmont 1:fdd22bb7aa52 170 }
emilmont 1:fdd22bb7aa52 171
emilmont 1:fdd22bb7aa52 172 /**
emilmont 1:fdd22bb7aa52 173 * @} end of cmplx_conj group
emilmont 1:fdd22bb7aa52 174 */