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
|
/*
* Copyright (c) 2022 Arm Limited.
*
* SPDX-License-Identifier: MIT
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to
* deal in the Software without restriction, including without limitation the
* rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
* sell copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in all
* copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef ARM_COMPUTE_WRAPPER_SHR_H
#define ARM_COMPUTE_WRAPPER_SHR_H
#include <type_traits>
#include <arm_neon.h>
namespace arm_compute
{
namespace wrapper
{
#define VQRSHRN_IMPL(half_vtype, vtype, prefix, postfix) \
template <int b> \
inline half_vtype vqrshrn(const vtype &a) \
{ \
return prefix##_##postfix(a, b); \
}
VQRSHRN_IMPL(int8x8_t, int16x8_t, vqrshrn_n, s16)
VQRSHRN_IMPL(uint8x8_t, uint16x8_t, vqrshrn_n, u16)
VQRSHRN_IMPL(int16x4_t, int32x4_t, vqrshrn_n, s32)
VQRSHRN_IMPL(uint16x4_t, uint32x4_t, vqrshrn_n, u32)
VQRSHRN_IMPL(int32x2_t, int64x2_t, vqrshrn_n, s64)
VQRSHRN_IMPL(uint32x2_t, uint64x2_t, vqrshrn_n, u64)
#undef VQRSHRN_IMPL
#ifdef __aarch64__
#define VQRSHRN_SCALAR_IMPL(half_vtype, vtype, prefix, postfix) \
template <int b> \
inline half_vtype vqrshrn(const vtype &a) \
{ \
return prefix##_##postfix(a, b); \
}
VQRSHRN_SCALAR_IMPL(int8_t, int16_t, vqrshrnh_n, s16)
VQRSHRN_SCALAR_IMPL(uint8_t, uint16_t, vqrshrnh_n, u16)
VQRSHRN_SCALAR_IMPL(int16_t, int32_t, vqrshrns_n, s32)
VQRSHRN_SCALAR_IMPL(uint16_t, uint32_t, vqrshrns_n, u32)
VQRSHRN_SCALAR_IMPL(int32_t, int64_t, vqrshrnd_n, s64)
VQRSHRN_SCALAR_IMPL(uint32_t, uint64_t, vqrshrnd_n, u64)
#undef VQRSHRN_SCALAR_IMPL
#endif // __aarch64__
// This function is the mixed version of VQRSHRN and VQRSHRUN.
// The input vector is always signed integer, while the returned vector
// can be either signed or unsigned depending on the signedness of scalar type T.
#define VQRSHRN_EX_IMPL(half_vtype, vtype, prefix_signed, prefix_unsigned, postfix) \
template <int b, typename T> \
inline typename std::enable_if<std::is_integral<T>::value && std::is_signed<T>::value, half_vtype>::type \
vqrshrn_ex(const vtype &a) \
{ \
return prefix_signed##_##postfix(a, b); \
} \
\
template <int b, typename T> \
inline typename std::enable_if<std::is_integral<T>::value && !std::is_signed<T>::value, u##half_vtype>::type \
vqrshrn_ex(const vtype &a) \
{ \
return prefix_unsigned##_##postfix(a, b); \
}
VQRSHRN_EX_IMPL(int8x8_t, int16x8_t, vqrshrn_n, vqrshrun_n, s16)
VQRSHRN_EX_IMPL(int16x4_t, int32x4_t, vqrshrn_n, vqrshrun_n, s32)
VQRSHRN_EX_IMPL(int32x2_t, int64x2_t, vqrshrn_n, vqrshrun_n, s64)
#undef VQRSHRN_EX_IMPL
#ifdef __aarch64__
#define VQRSHRN_EX_SCALAR_IMPL(half_vtype, vtype, prefix_signed, prefix_unsigned, postfix) \
template <int b, typename T> \
inline typename std::enable_if<std::is_integral<T>::value && std::is_signed<T>::value, half_vtype>::type \
vqrshrn_ex(const vtype &a) \
{ \
return prefix_signed##_##postfix(a, b); \
} \
\
template <int b, typename T> \
inline typename std::enable_if<std::is_integral<T>::value && !std::is_signed<T>::value, u##half_vtype>::type \
vqrshrn_ex(const vtype &a) \
{ \
return prefix_unsigned##_##postfix(a, b); \
}
VQRSHRN_EX_SCALAR_IMPL(int8_t, int16_t, vqrshrnh_n, vqrshrunh_n, s16)
VQRSHRN_EX_SCALAR_IMPL(int16_t, int32_t, vqrshrns_n, vqrshruns_n, s32)
VQRSHRN_EX_SCALAR_IMPL(int32_t, int64_t, vqrshrnd_n, vqrshrund_n, s64)
#undef VQRSHRN_EX_IMPL
#endif // __aarch64__
} // namespace wrapper
} // namespace arm_compute
#endif /* ARM_COMPUTE_WRAPPER_SHR_H */
|
