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_q15.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_cmplx_conj_q15.c
emilmont 1:fdd22bb7aa52 9 *
emilmont 1:fdd22bb7aa52 10 * Description: Q15 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
emilmont 1:fdd22bb7aa52 33 #include "arm_math.h"
emilmont 1:fdd22bb7aa52 34
emilmont 1:fdd22bb7aa52 35 /**
emilmont 1:fdd22bb7aa52 36 * @ingroup groupCmplxMath
emilmont 1:fdd22bb7aa52 37 */
emilmont 1:fdd22bb7aa52 38
emilmont 1:fdd22bb7aa52 39 /**
emilmont 1:fdd22bb7aa52 40 * @addtogroup cmplx_conj
emilmont 1:fdd22bb7aa52 41 * @{
emilmont 1:fdd22bb7aa52 42 */
emilmont 1:fdd22bb7aa52 43
emilmont 1:fdd22bb7aa52 44 /**
emilmont 1:fdd22bb7aa52 45 * @brief Q15 complex conjugate.
emilmont 1:fdd22bb7aa52 46 * @param *pSrc points to the input vector
emilmont 1:fdd22bb7aa52 47 * @param *pDst points to the output vector
emilmont 1:fdd22bb7aa52 48 * @param numSamples number of complex samples in each vector
emilmont 1:fdd22bb7aa52 49 * @return none.
emilmont 1:fdd22bb7aa52 50 *
emilmont 1:fdd22bb7aa52 51 * <b>Scaling and Overflow Behavior:</b>
emilmont 1:fdd22bb7aa52 52 * \par
emilmont 1:fdd22bb7aa52 53 * The function uses saturating arithmetic.
emilmont 1:fdd22bb7aa52 54 * The Q15 value -1 (0x8000) will be saturated to the maximum allowable positive value 0x7FFF.
emilmont 1:fdd22bb7aa52 55 */
emilmont 1:fdd22bb7aa52 56
emilmont 1:fdd22bb7aa52 57 void arm_cmplx_conj_q15(
emilmont 1:fdd22bb7aa52 58 q15_t * pSrc,
emilmont 1:fdd22bb7aa52 59 q15_t * pDst,
emilmont 1:fdd22bb7aa52 60 uint32_t numSamples)
emilmont 1:fdd22bb7aa52 61 {
emilmont 1:fdd22bb7aa52 62
emilmont 1:fdd22bb7aa52 63 #ifndef ARM_MATH_CM0
emilmont 1:fdd22bb7aa52 64
emilmont 1:fdd22bb7aa52 65 /* Run the below code for Cortex-M4 and Cortex-M3 */
emilmont 1:fdd22bb7aa52 66 uint32_t blkCnt; /* loop counter */
emilmont 1:fdd22bb7aa52 67 q31_t in1, in2, in3, in4;
emilmont 1:fdd22bb7aa52 68 q31_t zero = 0;
emilmont 1:fdd22bb7aa52 69
emilmont 1:fdd22bb7aa52 70 /*loop Unrolling */
emilmont 1:fdd22bb7aa52 71 blkCnt = numSamples >> 2u;
emilmont 1:fdd22bb7aa52 72
emilmont 1:fdd22bb7aa52 73 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
emilmont 1:fdd22bb7aa52 74 ** a second loop below computes the remaining 1 to 3 samples. */
emilmont 1:fdd22bb7aa52 75 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 76 {
emilmont 1:fdd22bb7aa52 77 /* C[0]+jC[1] = A[0]+ j (-1) A[1] */
emilmont 1:fdd22bb7aa52 78 /* Calculate Complex Conjugate and then store the results in the destination buffer. */
emilmont 1:fdd22bb7aa52 79 in1 = *__SIMD32(pSrc)++;
emilmont 1:fdd22bb7aa52 80 in2 = *__SIMD32(pSrc)++;
emilmont 1:fdd22bb7aa52 81 in3 = *__SIMD32(pSrc)++;
emilmont 1:fdd22bb7aa52 82 in4 = *__SIMD32(pSrc)++;
emilmont 1:fdd22bb7aa52 83
emilmont 1:fdd22bb7aa52 84 #ifndef ARM_MATH_BIG_ENDIAN
emilmont 1:fdd22bb7aa52 85
emilmont 1:fdd22bb7aa52 86 in1 = __QASX(zero, in1);
emilmont 1:fdd22bb7aa52 87 in2 = __QASX(zero, in2);
emilmont 1:fdd22bb7aa52 88 in3 = __QASX(zero, in3);
emilmont 1:fdd22bb7aa52 89 in4 = __QASX(zero, in4);
emilmont 1:fdd22bb7aa52 90
emilmont 1:fdd22bb7aa52 91 #else
emilmont 1:fdd22bb7aa52 92
emilmont 1:fdd22bb7aa52 93 in1 = __QSAX(zero, in1);
emilmont 1:fdd22bb7aa52 94 in2 = __QSAX(zero, in2);
emilmont 1:fdd22bb7aa52 95 in3 = __QSAX(zero, in3);
emilmont 1:fdd22bb7aa52 96 in4 = __QSAX(zero, in4);
emilmont 1:fdd22bb7aa52 97
emilmont 1:fdd22bb7aa52 98 #endif // #ifndef ARM_MATH_BIG_ENDIAN
emilmont 1:fdd22bb7aa52 99
emilmont 1:fdd22bb7aa52 100 in1 = ((uint32_t) in1 >> 16) | ((uint32_t) in1 << 16);
emilmont 1:fdd22bb7aa52 101 in2 = ((uint32_t) in2 >> 16) | ((uint32_t) in2 << 16);
emilmont 1:fdd22bb7aa52 102 in3 = ((uint32_t) in3 >> 16) | ((uint32_t) in3 << 16);
emilmont 1:fdd22bb7aa52 103 in4 = ((uint32_t) in4 >> 16) | ((uint32_t) in4 << 16);
emilmont 1:fdd22bb7aa52 104
emilmont 1:fdd22bb7aa52 105 *__SIMD32(pDst)++ = in1;
emilmont 1:fdd22bb7aa52 106 *__SIMD32(pDst)++ = in2;
emilmont 1:fdd22bb7aa52 107 *__SIMD32(pDst)++ = in3;
emilmont 1:fdd22bb7aa52 108 *__SIMD32(pDst)++ = in4;
emilmont 1:fdd22bb7aa52 109
emilmont 1:fdd22bb7aa52 110 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 111 blkCnt--;
emilmont 1:fdd22bb7aa52 112 }
emilmont 1:fdd22bb7aa52 113
emilmont 1:fdd22bb7aa52 114 /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
emilmont 1:fdd22bb7aa52 115 ** No loop unrolling is used. */
emilmont 1:fdd22bb7aa52 116 blkCnt = numSamples % 0x4u;
emilmont 1:fdd22bb7aa52 117
emilmont 1:fdd22bb7aa52 118 while(blkCnt > 0u)
emilmont 1:fdd22bb7aa52 119 {
emilmont 1:fdd22bb7aa52 120 /* C[0]+jC[1] = A[0]+ j (-1) A[1] */
emilmont 1:fdd22bb7aa52 121 /* Calculate Complex Conjugate and then store the results in the destination buffer. */
emilmont 1:fdd22bb7aa52 122 *pDst++ = *pSrc++;
emilmont 1:fdd22bb7aa52 123 *pDst++ = __SSAT(-*pSrc++, 16);
emilmont 1:fdd22bb7aa52 124
emilmont 1:fdd22bb7aa52 125 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 126 blkCnt--;
emilmont 1:fdd22bb7aa52 127 }
emilmont 1:fdd22bb7aa52 128
emilmont 1:fdd22bb7aa52 129 #else
emilmont 1:fdd22bb7aa52 130
emilmont 1:fdd22bb7aa52 131 q15_t in;
emilmont 1:fdd22bb7aa52 132
emilmont 1:fdd22bb7aa52 133 /* Run the below code for Cortex-M0 */
emilmont 1:fdd22bb7aa52 134
emilmont 1:fdd22bb7aa52 135 while(numSamples > 0u)
emilmont 1:fdd22bb7aa52 136 {
emilmont 1:fdd22bb7aa52 137 /* realOut + j (imagOut) = realIn+ j (-1) imagIn */
emilmont 1:fdd22bb7aa52 138 /* Calculate Complex Conjugate and then store the results in the destination buffer. */
emilmont 1:fdd22bb7aa52 139 *pDst++ = *pSrc++;
emilmont 1:fdd22bb7aa52 140 in = *pSrc++;
emilmont 1:fdd22bb7aa52 141 *pDst++ = (in == (q15_t) 0x8000) ? 0x7fff : -in;
emilmont 1:fdd22bb7aa52 142
emilmont 1:fdd22bb7aa52 143 /* Decrement the loop counter */
emilmont 1:fdd22bb7aa52 144 numSamples--;
emilmont 1:fdd22bb7aa52 145 }
emilmont 1:fdd22bb7aa52 146
emilmont 1:fdd22bb7aa52 147 #endif /* #ifndef ARM_MATH_CM0 */
emilmont 1:fdd22bb7aa52 148
emilmont 1:fdd22bb7aa52 149 }
emilmont 1:fdd22bb7aa52 150
emilmont 1:fdd22bb7aa52 151 /**
emilmont 1:fdd22bb7aa52 152 * @} end of cmplx_conj group
emilmont 1:fdd22bb7aa52 153 */