1 /* ----------------------------------------------------------------------
2 * Project: CMSIS DSP Library
3 * Title: arm_accumulate_f64.c
4 * Description: Accumulation value of a floating-point vector
5 *
6 * $Date: 14 July 2022
7 * $Revision: V1.0.0
8 *
9 * Target Processor: Cortex-M and Cortex-A cores
10 * -------------------------------------------------------------------- */
11 /*
12 * Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
13 *
14 * SPDX-License-Identifier: Apache-2.0
15 *
16 * Licensed under the Apache License, Version 2.0 (the License); you may
17 * not use this file except in compliance with the License.
18 * You may obtain a copy of the License at
19 *
20 * www.apache.org/licenses/LICENSE-2.0
21 *
22 * Unless required by applicable law or agreed to in writing, software
23 * distributed under the License is distributed on an AS IS BASIS, WITHOUT
24 * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
25 * See the License for the specific language governing permissions and
26 * limitations under the License.
27 */
28
29 #include "dsp/statistics_functions.h"
30
31 /**
32 @ingroup groupStats
33 */
34
35
36 /**
37 @addtogroup Accumulation
38 @{
39 */
40
41 /**
42 @brief Accumulation value of a floating-point vector.
43 @param[in] pSrc points to the input vector.
44 @param[in] blockSize number of samples in input vector.
45 @param[out] pResult sum of values in input vector.
46 */
47 #if defined(ARM_MATH_NEON) && defined(__aarch64__)
arm_accumulate_f64(const float64_t * pSrc,uint32_t blockSize,float64_t * pResult)48 ARM_DSP_ATTRIBUTE void arm_accumulate_f64(
49 const float64_t * pSrc,
50 uint32_t blockSize,
51 float64_t * pResult)
52 {
53 uint32_t blkCnt; /* Loop counter */
54
55 /*Neon buffers*/
56 float64x2_t vSum = vdupq_n_f64(0.0);
57 float64x2_t afterLoad ;
58
59 float64_t sum = 0.; /* Temporary result storage */
60
61 /* Initialize blkCnt with number of samples */
62 blkCnt = blockSize >> 1U;
63
64
65 while (blkCnt > 0U)
66 {
67 /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) */
68
69 afterLoad = vld1q_f64(pSrc);
70 vSum = vaddq_f64(vSum, afterLoad);
71
72 /* Decrement loop counter */
73 blkCnt--;
74
75 pSrc += 2;
76 }
77 sum = vaddvq_f64(vSum);
78
79 /* Tail */
80 blkCnt = blockSize & 1 ;
81
82 while (blkCnt > 0U)
83 {
84 /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) */
85 sum += *pSrc++;
86
87 /* Decrement loop counter */
88 blkCnt--;
89 }
90
91 /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) */
92 /* Store result to destination */
93 *pResult = sum;
94 }
95 #else
arm_accumulate_f64(const float64_t * pSrc,uint32_t blockSize,float64_t * pResult)96 ARM_DSP_ATTRIBUTE void arm_accumulate_f64(
97 const float64_t * pSrc,
98 uint32_t blockSize,
99 float64_t * pResult)
100 {
101 uint32_t blkCnt; /* Loop counter */
102 float64_t sum = 0.; /* Temporary result storage */
103
104 /* Initialize blkCnt with number of samples */
105 blkCnt = blockSize;
106
107 while (blkCnt > 0U)
108 {
109 /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) */
110 sum += *pSrc++;
111
112 /* Decrement loop counter */
113 blkCnt--;
114 }
115
116 /* C = (A[0] + A[1] + A[2] + ... + A[blockSize-1]) */
117 /* Store result to destination */
118 *pResult = sum;
119 }
120
121 #endif
122
123
124 /**
125 @} end of Accumulation group
126 */
127
