blob: a00435394fd808ed6e165f4df8152fc2a7cbe659 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
|
/* ----------------------------------------------------------------------
* Project: CMSIS DSP Library
* Title: arm_float_to_q15.c
* Description: Converts the elements of the floating-point vector to Q15 vector
*
* $Date: 23 April 2021
* $Revision: V1.9.0
*
* Target Processor: Cortex-M and Cortex-A cores
* -------------------------------------------------------------------- */
/*
* Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
*
* SPDX-License-Identifier: Apache-2.0
*
* Licensed under the Apache License, Version 2.0 (the License); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an AS IS BASIS, WITHOUT
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include "dsp/support_functions_f16.h"
#if defined(ARM_FLOAT16_SUPPORTED)
/**
@ingroup groupSupport
*/
/**
* @defgroup f16_to_x Convert 16-bit floating point value
*/
/**
@addtogroup f16_to_x
@{
*/
/**
@brief Converts the elements of the f16 vector to f32 vector.
@param[in] pSrc points to the f16 input vector
@param[out] pDst points to the f32 output vector
@param[in] blockSize number of samples in each vector
@return none
*/
#if defined(ARM_MATH_MVE_FLOAT16) && !defined(ARM_MATH_AUTOVECTORIZE) && defined(__CMSIS_GCC_H)
#pragma GCC warning "Scalar version of arm_f16_to_float built. Helium version has build issues with gcc."
#endif
#if defined(ARM_MATH_MVE_FLOAT16) && !defined(ARM_MATH_AUTOVECTORIZE) && !defined(__CMSIS_GCC_H)
void arm_f16_to_float(
const float16_t * pSrc,
float32_t * pDst,
uint32_t blockSize)
{
int32_t blkCnt; /* loop counters */
float16x8_t vecDst;
float32x4x2_t tmp;
blkCnt = blockSize >> 3;
while (blkCnt > 0)
{
vecDst = vldrhq_f16(pSrc);
pSrc += 8;
tmp.val[0] = vcvtbq_f32_f16(vecDst);
tmp.val[1] = vcvttq_f32_f16(vecDst);
vst2q(pDst,tmp);
pDst += 8;
/*
* Decrement the blockSize loop counter
*/
blkCnt--;
}
/*
* tail
* (will be merged thru tail predication)
*/
blkCnt = blockSize & 7;
while (blkCnt > 0)
{
*pDst++ = (float32_t) *pSrc++;
/*
* Decrement the loop counter
*/
blkCnt--;
}
}
#else
void arm_f16_to_float(
const float16_t * pSrc,
float32_t * pDst,
uint32_t blockSize)
{
const float16_t *pIn = pSrc; /* Src pointer */
uint32_t blkCnt; /* loop counter */
/*
* Loop over blockSize number of values
*/
blkCnt = blockSize;
while (blkCnt > 0U)
{
*pDst++ = (float32_t) * pIn++;
/*
* Decrement the loop counter
*/
blkCnt--;
}
}
#endif /* defined(ARM_MATH_MVEF) && !defined(ARM_MATH_AUTOVECTORIZE) */
/**
@} end of f16_to_x group
*/
#endif /* #if defined(ARM_FLOAT16_SUPPORTED) */
|
