Fork of mbed-dsp. CMSIS-DSP library of supporting NEON
Dependents: mbed-os-example-cmsis_dsp_neon
Fork of mbed-dsp by
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/StatisticsFunctions/arm_std_q15.c@4:9cee975aadce, 2014-06-23 (annotated)
- Committer:
- mbed_official
- Date:
- Mon Jun 23 09:30:09 2014 +0100
- Revision:
- 4:9cee975aadce
- Parent:
- 3:7a284390b0ce
Synchronized with git revision 6e7c7bcec41226f536474daae3c13d49e4c0e865
Full URL: https://github.com/mbedmicro/mbed/commit/6e7c7bcec41226f536474daae3c13d49e4c0e865/
Fix signed unsigned compare in dsp library
Who changed what in which revision?
User | Revision | Line number | New contents of line |
---|---|---|---|
emilmont | 1:fdd22bb7aa52 | 1 | /* ---------------------------------------------------------------------- |
mbed_official | 3:7a284390b0ce | 2 | * Copyright (C) 2010-2013 ARM Limited. All rights reserved. |
emilmont | 1:fdd22bb7aa52 | 3 | * |
mbed_official | 3:7a284390b0ce | 4 | * $Date: 17. January 2013 |
mbed_official | 3:7a284390b0ce | 5 | * $Revision: V1.4.1 |
emilmont | 1:fdd22bb7aa52 | 6 | * |
emilmont | 2:da51fb522205 | 7 | * Project: CMSIS DSP Library |
emilmont | 2:da51fb522205 | 8 | * Title: arm_std_q15.c |
emilmont | 1:fdd22bb7aa52 | 9 | * |
emilmont | 2:da51fb522205 | 10 | * Description: Standard deviation of an array of Q15 type. |
emilmont | 1:fdd22bb7aa52 | 11 | * |
emilmont | 1:fdd22bb7aa52 | 12 | * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0 |
emilmont | 1:fdd22bb7aa52 | 13 | * |
mbed_official | 3:7a284390b0ce | 14 | * Redistribution and use in source and binary forms, with or without |
mbed_official | 3:7a284390b0ce | 15 | * modification, are permitted provided that the following conditions |
mbed_official | 3:7a284390b0ce | 16 | * are met: |
mbed_official | 3:7a284390b0ce | 17 | * - Redistributions of source code must retain the above copyright |
mbed_official | 3:7a284390b0ce | 18 | * notice, this list of conditions and the following disclaimer. |
mbed_official | 3:7a284390b0ce | 19 | * - Redistributions in binary form must reproduce the above copyright |
mbed_official | 3:7a284390b0ce | 20 | * notice, this list of conditions and the following disclaimer in |
mbed_official | 3:7a284390b0ce | 21 | * the documentation and/or other materials provided with the |
mbed_official | 3:7a284390b0ce | 22 | * distribution. |
mbed_official | 3:7a284390b0ce | 23 | * - Neither the name of ARM LIMITED nor the names of its contributors |
mbed_official | 3:7a284390b0ce | 24 | * may be used to endorse or promote products derived from this |
mbed_official | 3:7a284390b0ce | 25 | * software without specific prior written permission. |
mbed_official | 3:7a284390b0ce | 26 | * |
mbed_official | 3:7a284390b0ce | 27 | * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
mbed_official | 3:7a284390b0ce | 28 | * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
mbed_official | 3:7a284390b0ce | 29 | * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS |
mbed_official | 3:7a284390b0ce | 30 | * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE |
mbed_official | 3:7a284390b0ce | 31 | * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, |
mbed_official | 3:7a284390b0ce | 32 | * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, |
mbed_official | 3:7a284390b0ce | 33 | * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; |
mbed_official | 3:7a284390b0ce | 34 | * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER |
mbed_official | 3:7a284390b0ce | 35 | * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
mbed_official | 3:7a284390b0ce | 36 | * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN |
mbed_official | 3:7a284390b0ce | 37 | * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE |
mbed_official | 3:7a284390b0ce | 38 | * POSSIBILITY OF SUCH DAMAGE. |
emilmont | 1:fdd22bb7aa52 | 39 | * -------------------------------------------------------------------- */ |
emilmont | 1:fdd22bb7aa52 | 40 | |
emilmont | 1:fdd22bb7aa52 | 41 | #include "arm_math.h" |
emilmont | 1:fdd22bb7aa52 | 42 | |
emilmont | 1:fdd22bb7aa52 | 43 | /** |
emilmont | 1:fdd22bb7aa52 | 44 | * @ingroup groupStats |
emilmont | 1:fdd22bb7aa52 | 45 | */ |
emilmont | 1:fdd22bb7aa52 | 46 | |
emilmont | 1:fdd22bb7aa52 | 47 | /** |
emilmont | 1:fdd22bb7aa52 | 48 | * @addtogroup STD |
emilmont | 1:fdd22bb7aa52 | 49 | * @{ |
emilmont | 1:fdd22bb7aa52 | 50 | */ |
emilmont | 1:fdd22bb7aa52 | 51 | |
emilmont | 1:fdd22bb7aa52 | 52 | /** |
emilmont | 1:fdd22bb7aa52 | 53 | * @brief Standard deviation of the elements of a Q15 vector. |
emilmont | 1:fdd22bb7aa52 | 54 | * @param[in] *pSrc points to the input vector |
emilmont | 1:fdd22bb7aa52 | 55 | * @param[in] blockSize length of the input vector |
emilmont | 1:fdd22bb7aa52 | 56 | * @param[out] *pResult standard deviation value returned here |
emilmont | 1:fdd22bb7aa52 | 57 | * @return none. |
emilmont | 1:fdd22bb7aa52 | 58 | * |
emilmont | 1:fdd22bb7aa52 | 59 | * @details |
emilmont | 1:fdd22bb7aa52 | 60 | * <b>Scaling and Overflow Behavior:</b> |
emilmont | 1:fdd22bb7aa52 | 61 | * |
emilmont | 1:fdd22bb7aa52 | 62 | * \par |
emilmont | 1:fdd22bb7aa52 | 63 | * The function is implemented using a 64-bit internal accumulator. |
emilmont | 1:fdd22bb7aa52 | 64 | * The input is represented in 1.15 format. |
emilmont | 1:fdd22bb7aa52 | 65 | * Intermediate multiplication yields a 2.30 format, and this |
emilmont | 1:fdd22bb7aa52 | 66 | * result is added without saturation to a 64-bit accumulator in 34.30 format. |
emilmont | 1:fdd22bb7aa52 | 67 | * With 33 guard bits in the accumulator, there is no risk of overflow, and the |
emilmont | 1:fdd22bb7aa52 | 68 | * full precision of the intermediate multiplication is preserved. |
emilmont | 1:fdd22bb7aa52 | 69 | * Finally, the 34.30 result is truncated to 34.15 format by discarding the lower |
emilmont | 1:fdd22bb7aa52 | 70 | * 15 bits, and then saturated to yield a result in 1.15 format. |
emilmont | 1:fdd22bb7aa52 | 71 | */ |
emilmont | 1:fdd22bb7aa52 | 72 | |
emilmont | 1:fdd22bb7aa52 | 73 | void arm_std_q15( |
emilmont | 1:fdd22bb7aa52 | 74 | q15_t * pSrc, |
emilmont | 1:fdd22bb7aa52 | 75 | uint32_t blockSize, |
emilmont | 1:fdd22bb7aa52 | 76 | q15_t * pResult) |
emilmont | 1:fdd22bb7aa52 | 77 | { |
emilmont | 1:fdd22bb7aa52 | 78 | q31_t sum = 0; /* Accumulator */ |
emilmont | 1:fdd22bb7aa52 | 79 | q31_t meanOfSquares, squareOfMean; /* square of mean and mean of square */ |
emilmont | 1:fdd22bb7aa52 | 80 | q15_t mean; /* mean */ |
emilmont | 1:fdd22bb7aa52 | 81 | uint32_t blkCnt; /* loop counter */ |
emilmont | 1:fdd22bb7aa52 | 82 | q15_t t; /* Temporary variable */ |
emilmont | 1:fdd22bb7aa52 | 83 | q63_t sumOfSquares = 0; /* Accumulator */ |
emilmont | 1:fdd22bb7aa52 | 84 | |
mbed_official | 3:7a284390b0ce | 85 | #ifndef ARM_MATH_CM0_FAMILY |
emilmont | 1:fdd22bb7aa52 | 86 | |
emilmont | 1:fdd22bb7aa52 | 87 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
emilmont | 1:fdd22bb7aa52 | 88 | |
emilmont | 1:fdd22bb7aa52 | 89 | q31_t in; /* input value */ |
emilmont | 1:fdd22bb7aa52 | 90 | q15_t in1; /* input value */ |
emilmont | 1:fdd22bb7aa52 | 91 | |
emilmont | 1:fdd22bb7aa52 | 92 | /*loop Unrolling */ |
emilmont | 1:fdd22bb7aa52 | 93 | blkCnt = blockSize >> 2u; |
emilmont | 1:fdd22bb7aa52 | 94 | |
emilmont | 1:fdd22bb7aa52 | 95 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
emilmont | 1:fdd22bb7aa52 | 96 | ** a second loop below computes the remaining 1 to 3 samples. */ |
emilmont | 1:fdd22bb7aa52 | 97 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 98 | { |
emilmont | 1:fdd22bb7aa52 | 99 | /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */ |
emilmont | 1:fdd22bb7aa52 | 100 | /* Compute Sum of squares of the input samples |
emilmont | 1:fdd22bb7aa52 | 101 | * and then store the result in a temporary variable, sum. */ |
emilmont | 1:fdd22bb7aa52 | 102 | in = *__SIMD32(pSrc)++; |
emilmont | 1:fdd22bb7aa52 | 103 | sum += ((in << 16) >> 16); |
emilmont | 1:fdd22bb7aa52 | 104 | sum += (in >> 16); |
emilmont | 1:fdd22bb7aa52 | 105 | sumOfSquares = __SMLALD(in, in, sumOfSquares); |
emilmont | 1:fdd22bb7aa52 | 106 | in = *__SIMD32(pSrc)++; |
emilmont | 1:fdd22bb7aa52 | 107 | sum += ((in << 16) >> 16); |
emilmont | 1:fdd22bb7aa52 | 108 | sum += (in >> 16); |
emilmont | 1:fdd22bb7aa52 | 109 | sumOfSquares = __SMLALD(in, in, sumOfSquares); |
emilmont | 1:fdd22bb7aa52 | 110 | |
emilmont | 1:fdd22bb7aa52 | 111 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 112 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 113 | } |
emilmont | 1:fdd22bb7aa52 | 114 | |
emilmont | 1:fdd22bb7aa52 | 115 | /* If the blockSize is not a multiple of 4, compute any remaining output samples here. |
emilmont | 1:fdd22bb7aa52 | 116 | ** No loop unrolling is used. */ |
emilmont | 1:fdd22bb7aa52 | 117 | blkCnt = blockSize % 0x4u; |
emilmont | 1:fdd22bb7aa52 | 118 | |
emilmont | 1:fdd22bb7aa52 | 119 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 120 | { |
emilmont | 1:fdd22bb7aa52 | 121 | /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */ |
emilmont | 1:fdd22bb7aa52 | 122 | /* Compute Sum of squares of the input samples |
emilmont | 1:fdd22bb7aa52 | 123 | * and then store the result in a temporary variable, sum. */ |
emilmont | 1:fdd22bb7aa52 | 124 | in1 = *pSrc++; |
emilmont | 1:fdd22bb7aa52 | 125 | sumOfSquares = __SMLALD(in1, in1, sumOfSquares); |
emilmont | 1:fdd22bb7aa52 | 126 | sum += in1; |
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 | /* Compute Mean of squares of the input samples |
emilmont | 1:fdd22bb7aa52 | 133 | * and then store the result in a temporary variable, meanOfSquares. */ |
emilmont | 1:fdd22bb7aa52 | 134 | t = (q15_t) ((1.0 / (blockSize - 1)) * 16384LL); |
emilmont | 1:fdd22bb7aa52 | 135 | sumOfSquares = __SSAT((sumOfSquares >> 15u), 16u); |
emilmont | 1:fdd22bb7aa52 | 136 | |
emilmont | 1:fdd22bb7aa52 | 137 | meanOfSquares = (q31_t) ((sumOfSquares * t) >> 14u); |
emilmont | 1:fdd22bb7aa52 | 138 | |
emilmont | 1:fdd22bb7aa52 | 139 | /* Compute mean of all input values */ |
emilmont | 1:fdd22bb7aa52 | 140 | t = (q15_t) ((1.0 / (blockSize * (blockSize - 1))) * 32768LL); |
emilmont | 1:fdd22bb7aa52 | 141 | mean = (q15_t) __SSAT(sum, 16u); |
emilmont | 1:fdd22bb7aa52 | 142 | |
emilmont | 1:fdd22bb7aa52 | 143 | /* Compute square of mean */ |
emilmont | 1:fdd22bb7aa52 | 144 | squareOfMean = ((q31_t) mean * mean) >> 15; |
emilmont | 1:fdd22bb7aa52 | 145 | squareOfMean = (q31_t) (((q63_t) squareOfMean * t) >> 15); |
emilmont | 1:fdd22bb7aa52 | 146 | |
emilmont | 1:fdd22bb7aa52 | 147 | /* mean of the squares minus the square of the mean. */ |
emilmont | 1:fdd22bb7aa52 | 148 | in1 = (q15_t) (meanOfSquares - squareOfMean); |
emilmont | 1:fdd22bb7aa52 | 149 | |
emilmont | 1:fdd22bb7aa52 | 150 | /* Compute standard deviation and store the result to the destination */ |
emilmont | 1:fdd22bb7aa52 | 151 | arm_sqrt_q15(in1, pResult); |
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 | q15_t in; /* input value */ |
emilmont | 1:fdd22bb7aa52 | 157 | |
emilmont | 1:fdd22bb7aa52 | 158 | /* Loop over blockSize number of values */ |
emilmont | 1:fdd22bb7aa52 | 159 | blkCnt = blockSize; |
emilmont | 1:fdd22bb7aa52 | 160 | |
emilmont | 1:fdd22bb7aa52 | 161 | while(blkCnt > 0u) |
emilmont | 1:fdd22bb7aa52 | 162 | { |
emilmont | 1:fdd22bb7aa52 | 163 | /* C = (A[0] * A[0] + A[1] * A[1] + ... + A[blockSize-1] * A[blockSize-1]) */ |
emilmont | 1:fdd22bb7aa52 | 164 | /* Compute Sum of squares of the input samples |
emilmont | 1:fdd22bb7aa52 | 165 | * and then store the result in a temporary variable, sumOfSquares. */ |
emilmont | 1:fdd22bb7aa52 | 166 | in = *pSrc++; |
emilmont | 1:fdd22bb7aa52 | 167 | sumOfSquares += (in * in); |
emilmont | 1:fdd22bb7aa52 | 168 | |
emilmont | 1:fdd22bb7aa52 | 169 | /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) */ |
emilmont | 1:fdd22bb7aa52 | 170 | /* Compute sum of all input values and then store the result in a temporary variable, sum. */ |
emilmont | 1:fdd22bb7aa52 | 171 | sum += in; |
emilmont | 1:fdd22bb7aa52 | 172 | |
emilmont | 1:fdd22bb7aa52 | 173 | /* Decrement the loop counter */ |
emilmont | 1:fdd22bb7aa52 | 174 | blkCnt--; |
emilmont | 1:fdd22bb7aa52 | 175 | } |
emilmont | 1:fdd22bb7aa52 | 176 | |
emilmont | 1:fdd22bb7aa52 | 177 | /* Compute Mean of squares of the input samples |
emilmont | 1:fdd22bb7aa52 | 178 | * and then store the result in a temporary variable, meanOfSquares. */ |
emilmont | 1:fdd22bb7aa52 | 179 | t = (q15_t) ((1.0 / (blockSize - 1)) * 16384LL); |
emilmont | 1:fdd22bb7aa52 | 180 | sumOfSquares = __SSAT((sumOfSquares >> 15u), 16u); |
emilmont | 1:fdd22bb7aa52 | 181 | meanOfSquares = (q31_t) ((sumOfSquares * t) >> 14u); |
emilmont | 1:fdd22bb7aa52 | 182 | |
emilmont | 1:fdd22bb7aa52 | 183 | /* Compute mean of all input values */ |
emilmont | 1:fdd22bb7aa52 | 184 | mean = (q15_t) __SSAT(sum, 16u); |
emilmont | 1:fdd22bb7aa52 | 185 | |
emilmont | 1:fdd22bb7aa52 | 186 | /* Compute square of mean of the input samples |
emilmont | 1:fdd22bb7aa52 | 187 | * and then store the result in a temporary variable, squareOfMean.*/ |
emilmont | 1:fdd22bb7aa52 | 188 | t = (q15_t) ((1.0 / (blockSize * (blockSize - 1))) * 32768LL); |
emilmont | 1:fdd22bb7aa52 | 189 | squareOfMean = ((q31_t) mean * mean) >> 15; |
emilmont | 1:fdd22bb7aa52 | 190 | squareOfMean = (q31_t) (((q63_t) squareOfMean * t) >> 15); |
emilmont | 1:fdd22bb7aa52 | 191 | |
emilmont | 1:fdd22bb7aa52 | 192 | /* mean of the squares minus the square of the mean. */ |
emilmont | 1:fdd22bb7aa52 | 193 | in = (q15_t) (meanOfSquares - squareOfMean); |
emilmont | 1:fdd22bb7aa52 | 194 | |
emilmont | 1:fdd22bb7aa52 | 195 | /* Compute standard deviation and store the result to the destination */ |
emilmont | 1:fdd22bb7aa52 | 196 | arm_sqrt_q15(in, pResult); |
emilmont | 1:fdd22bb7aa52 | 197 | |
mbed_official | 3:7a284390b0ce | 198 | #endif /* #ifndef ARM_MATH_CM0_FAMILY */ |
emilmont | 1:fdd22bb7aa52 | 199 | |
emilmont | 1:fdd22bb7aa52 | 200 | |
emilmont | 1:fdd22bb7aa52 | 201 | } |
emilmont | 1:fdd22bb7aa52 | 202 | |
emilmont | 1:fdd22bb7aa52 | 203 | /** |
emilmont | 1:fdd22bb7aa52 | 204 | * @} end of STD group |
emilmont | 1:fdd22bb7aa52 | 205 | */ |