forked from Tencent/ncnn
/
swish_arm_asimdhp.cpp
165 lines (149 loc) · 5.61 KB
/
swish_arm_asimdhp.cpp
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
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
// Tencent is pleased to support the open source community by making ncnn available.
//
// Copyright (C) 2022 THL A29 Limited, a Tencent company. All rights reserved.
//
// Licensed under the BSD 3-Clause License (the "License"); you may not use this file except
// in compliance with the License. You may obtain a copy of the License at
//
// https://opensource.org/licenses/BSD-3-Clause
//
// 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 "swish_arm.h"
#if __ARM_NEON
#include <arm_neon.h>
#include "neon_mathfun.h"
#if __ARM_FEATURE_FP16_VECTOR_ARITHMETIC
#include "neon_mathfun_fp16s.h"
#endif
#endif // __ARM_NEON
namespace ncnn {
#if __ARM_FEATURE_FP16_VECTOR_ARITHMETIC
int Swish_arm::forward_inplace_fp16s(Mat& bottom_top_blob, const Option& opt) const
{
int w = bottom_top_blob.w;
int h = bottom_top_blob.h;
int d = bottom_top_blob.d;
int channels = bottom_top_blob.c;
int elempack = bottom_top_blob.elempack;
int size = w * h * d * elempack;
#pragma omp parallel for num_threads(opt.num_threads)
for (int q = 0; q < channels; q++)
{
__fp16* ptr = bottom_top_blob.channel(q);
float32x4_t _one = vdupq_n_f32(1.f);
int i = 0;
for (; i + 15 < size; i += 16)
{
float16x8_t _p01 = vld1q_f16(ptr);
float16x8_t _p23 = vld1q_f16(ptr + 8);
float32x4_t _p0 = vcvt_f32_f16(vget_low_f16(_p01));
float32x4_t _p1 = vcvt_f32_f16(vget_high_f16(_p01));
float32x4_t _p2 = vcvt_f32_f16(vget_low_f16(_p23));
float32x4_t _p3 = vcvt_f32_f16(vget_high_f16(_p23));
_p0 = vdivq_f32(_p0, vaddq_f32(_one, exp_ps(vnegq_f32(_p0))));
_p1 = vdivq_f32(_p1, vaddq_f32(_one, exp_ps(vnegq_f32(_p1))));
_p2 = vdivq_f32(_p2, vaddq_f32(_one, exp_ps(vnegq_f32(_p2))));
_p3 = vdivq_f32(_p3, vaddq_f32(_one, exp_ps(vnegq_f32(_p3))));
_p01 = vcombine_f16(vcvt_f16_f32(_p0), vcvt_f16_f32(_p1));
_p23 = vcombine_f16(vcvt_f16_f32(_p2), vcvt_f16_f32(_p3));
vst1q_f16(ptr, _p01);
vst1q_f16(ptr + 8, _p23);
ptr += 16;
}
for (; i + 7 < size; i += 8)
{
float16x8_t _p = vld1q_f16(ptr);
float32x4_t _p0 = vcvt_f32_f16(vget_low_f16(_p));
float32x4_t _p1 = vcvt_f32_f16(vget_high_f16(_p));
_p0 = vdivq_f32(_p0, vaddq_f32(_one, exp_ps(vnegq_f32(_p0))));
_p1 = vdivq_f32(_p1, vaddq_f32(_one, exp_ps(vnegq_f32(_p1))));
_p = vcombine_f16(vcvt_f16_f32(_p0), vcvt_f16_f32(_p1));
vst1q_f16(ptr, _p);
ptr += 8;
}
for (; i + 3 < size; i += 4)
{
float32x4_t _p = vcvt_f32_f16(vld1_f16(ptr));
_p = vdivq_f32(_p, vaddq_f32(_one, exp_ps(vnegq_f32(_p))));
vst1_f16(ptr, vcvt_f16_f32(_p));
ptr += 4;
}
for (; i < size; i++)
{
float v = (float)*ptr;
v = v / (1.f + expf(-v));
*ptr = (__fp16)v;
ptr++;
}
}
return 0;
}
int Swish_arm::forward_inplace_fp16sa(Mat& bottom_top_blob, const Option& opt) const
{
int w = bottom_top_blob.w;
int h = bottom_top_blob.h;
int d = bottom_top_blob.d;
int channels = bottom_top_blob.c;
int elempack = bottom_top_blob.elempack;
int size = w * h * d * elempack;
#pragma omp parallel for num_threads(opt.num_threads)
for (int q = 0; q < channels; q++)
{
__fp16* ptr = bottom_top_blob.channel(q);
float16x8_t _one = vdupq_n_f16(1.f);
int i = 0;
for (; i + 31 < size; i += 32)
{
float16x8_t _p0 = vld1q_f16(ptr);
float16x8_t _p1 = vld1q_f16(ptr + 8);
float16x8_t _p2 = vld1q_f16(ptr + 16);
float16x8_t _p3 = vld1q_f16(ptr + 24);
_p0 = vdivq_f16(_p0, vaddq_f16(_one, exp_ps(vnegq_f16(_p0))));
_p1 = vdivq_f16(_p1, vaddq_f16(_one, exp_ps(vnegq_f16(_p1))));
_p2 = vdivq_f16(_p2, vaddq_f16(_one, exp_ps(vnegq_f16(_p2))));
_p3 = vdivq_f16(_p3, vaddq_f16(_one, exp_ps(vnegq_f16(_p3))));
vst1q_f16(ptr, _p0);
vst1q_f16(ptr + 8, _p1);
vst1q_f16(ptr + 16, _p2);
vst1q_f16(ptr + 24, _p3);
ptr += 32;
}
for (; i + 15 < size; i += 16)
{
float16x8_t _p0 = vld1q_f16(ptr);
float16x8_t _p1 = vld1q_f16(ptr + 8);
_p0 = vdivq_f16(_p0, vaddq_f16(_one, exp_ps(vnegq_f16(_p0))));
_p1 = vdivq_f16(_p1, vaddq_f16(_one, exp_ps(vnegq_f16(_p1))));
vst1q_f16(ptr, _p0);
vst1q_f16(ptr + 8, _p1);
ptr += 16;
}
for (; i + 7 < size; i += 8)
{
float16x8_t _p = vld1q_f16(ptr);
_p = vdivq_f16(_p, vaddq_f16(_one, exp_ps(vnegq_f16(_p))));
vst1q_f16(ptr, _p);
ptr += 8;
}
for (; i + 3 < size; i += 4)
{
float16x4_t _p = vld1_f16(ptr);
_p = vdiv_f16(_p, vadd_f16(vget_low_f16(_one), exp_ps(vneg_f16(_p))));
vst1_f16(ptr, _p);
ptr += 4;
}
for (; i < size; i++)
{
__fp16 v = *ptr;
v = v / ((__fp16)1.f + expf(-v));
*ptr = v;
ptr++;
}
}
return 0;
}
#endif // __ARM_FEATURE_FP16_VECTOR_ARITHMETIC
} // namespace ncnn