ncnn

1
// Tencent is pleased to support the open source community by making ncnn available.
2
//
3
// Copyright (C) 2020 THL A29 Limited, a Tencent company. All rights reserved.
4
//
5
// Licensed under the BSD 3-Clause License (the "License"); you may not use this file except
6
// in compliance with the License. You may obtain a copy of the License at
7
//
8
// https://opensource.org/licenses/BSD-3-Clause
9
//
10
// Unless required by applicable law or agreed to in writing, software distributed
11
// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR
12
// CONDITIONS OF ANY KIND, either express or implied. See the License for the
13
// specific language governing permissions and limitations under the License.
14

15
#version 450
16

17
#if NCNN_fp16_storage
18
#extension GL_EXT_shader_16bit_storage: require
19
struct sfpvec8 { f16vec4 abcd; f16vec4 efgh; };
20
#endif
21
#if NCNN_fp16_arithmetic
22
#extension GL_EXT_shader_explicit_arithmetic_types_float16: require
23
#endif
24

25
#extension GL_GOOGLE_include_directive: enable
26
#include "vulkan_activation.comp"
27

28
layout (constant_id = 0) const int bias_term = 0;
29
layout (constant_id = 1) const int activation_type = 0;
30
layout (constant_id = 2) const float activation_param_0 = 0;
31
layout (constant_id = 3) const float activation_param_1 = 0;
32

33
#define shape_constant_id_offset 4
34
layout (constant_id = shape_constant_id_offset + 0) const int dims = 0;
35
layout (constant_id = shape_constant_id_offset + 1) const int w = 0;
36
layout (constant_id = shape_constant_id_offset + 2) const int h = 0;
37
layout (constant_id = shape_constant_id_offset + 3) const int c = 0;
38
layout (constant_id = shape_constant_id_offset + 4) const int cstep = 0;
39

40
layout (constant_id = shape_constant_id_offset + 5) const int outdims = 0;
41
layout (constant_id = shape_constant_id_offset + 6) const int outw = 0;
42
layout (constant_id = shape_constant_id_offset + 7) const int outh = 0;
43
layout (constant_id = shape_constant_id_offset + 8) const int outc = 0;
44
layout (constant_id = shape_constant_id_offset + 9) const int outcstep = 0;
45

46
#if NCNN_image_shader
47
layout (binding = 0) uniform unfp sampler3D bottom_blob;
48
layout (binding = 1, imfmtc4) writeonly uniform unfp image3D top_blob;
49
layout (binding = 2) uniform unfp sampler3D weight_blob;
50
layout (binding = 3) uniform unfp sampler3D bias_blob;
51
#else
52
layout (binding = 0) readonly buffer bottom_blob { sfpvec8 bottom_blob_data[]; };
53
layout (binding = 1) writeonly buffer top_blob { sfpvec8 top_blob_data[]; };
54
layout (binding = 2) readonly buffer weight_blob { sfpvec8 weight_data[]; };
55
layout (binding = 3) readonly buffer bias_blob { sfpvec8 bias_data[]; };
56
#endif
57

58
layout (push_constant) uniform parameter
59
{
60
    int dims;
61
    int w;
62
    int h;
63
    int c;
64
    int cstep;
65

66
    int outdims;
67
    int outw;
68
    int outh;
69
    int outc;
70
    int outcstep;
71
} p;
72

73
void main()
74
{
75
    int gx = int(gl_GlobalInvocationID.x);
76
    int gy = int(gl_GlobalInvocationID.y);
77
    int gz = int(gl_GlobalInvocationID.z);
78

79
    if (gx >= psc(outw) || gy >= 1 || gz >= 1)
80
        return;
81

82
    afpvec8 sum;
83

84
    if (bias_term == 1)
85
    {
86
#if NCNN_image_shader
87
        sum = image3d_ld8(bias_blob, ivec3(gx, 0, 0));
88
#else
89
        sum = buffer_ld8(bias_data, gx);
90
#endif
91
    }
92
    else
93
    {
94
        sum = afpvec8(afpvec4(0.f), afpvec4(0.f));
95
    }
96

97
#if NCNN_image_shader
98
    int wx = 0;
99

100
    for (int i = 0; i < psc(w); i++)
101
    {
102
        afpvec8 v = image3d_ld8(bottom_blob, ivec3(i, 0, 0));
103

104
        afpvec8 k0 = image3d_ld8(weight_blob, ivec3(wx + 0, gx, 0));
105
        afpvec8 k1 = image3d_ld8(weight_blob, ivec3(wx + 1, gx, 0));
106
        afpvec8 k2 = image3d_ld8(weight_blob, ivec3(wx + 2, gx, 0));
107
        afpvec8 k3 = image3d_ld8(weight_blob, ivec3(wx + 3, gx, 0));
108
        afpvec8 k4 = image3d_ld8(weight_blob, ivec3(wx + 4, gx, 0));
109
        afpvec8 k5 = image3d_ld8(weight_blob, ivec3(wx + 5, gx, 0));
110
        afpvec8 k6 = image3d_ld8(weight_blob, ivec3(wx + 6, gx, 0));
111
        afpvec8 k7 = image3d_ld8(weight_blob, ivec3(wx + 7, gx, 0));
112

113
        // sum += v * k
114
        sum[0].r += dot(v[0], k0[0]) + dot(v[1], k0[1]);
115
        sum[0].g += dot(v[0], k1[0]) + dot(v[1], k1[1]);
116
        sum[0].b += dot(v[0], k2[0]) + dot(v[1], k2[1]);
117
        sum[0].a += dot(v[0], k3[0]) + dot(v[1], k3[1]);
118
        sum[1].r += dot(v[0], k4[0]) + dot(v[1], k4[1]);
119
        sum[1].g += dot(v[0], k5[0]) + dot(v[1], k5[1]);
120
        sum[1].b += dot(v[0], k6[0]) + dot(v[1], k6[1]);
121
        sum[1].a += dot(v[0], k7[0]) + dot(v[1], k7[1]);
122

123
        wx += 8;
124
    }
125
#else
126
    int w_offset = gx * psc(w);
127

128
    for (int i = 0; i < psc(w); i++)
129
    {
130
        afpvec8 v = buffer_ld8(bottom_blob_data, i);
131

132
        afpvec8 k0 = buffer_ld8(weight_data, (w_offset + i) * 8 + 0);
133
        afpvec8 k1 = buffer_ld8(weight_data, (w_offset + i) * 8 + 1);
134
        afpvec8 k2 = buffer_ld8(weight_data, (w_offset + i) * 8 + 2);
135
        afpvec8 k3 = buffer_ld8(weight_data, (w_offset + i) * 8 + 3);
136
        afpvec8 k4 = buffer_ld8(weight_data, (w_offset + i) * 8 + 4);
137
        afpvec8 k5 = buffer_ld8(weight_data, (w_offset + i) * 8 + 5);
138
        afpvec8 k6 = buffer_ld8(weight_data, (w_offset + i) * 8 + 6);
139
        afpvec8 k7 = buffer_ld8(weight_data, (w_offset + i) * 8 + 7);
140

141
        // sum += v * k
142
        sum[0].r += dot(v[0], k0[0]) + dot(v[1], k0[1]);
143
        sum[0].g += dot(v[0], k1[0]) + dot(v[1], k1[1]);
144
        sum[0].b += dot(v[0], k2[0]) + dot(v[1], k2[1]);
145
        sum[0].a += dot(v[0], k3[0]) + dot(v[1], k3[1]);
146
        sum[1].r += dot(v[0], k4[0]) + dot(v[1], k4[1]);
147
        sum[1].g += dot(v[0], k5[0]) + dot(v[1], k5[1]);
148
        sum[1].b += dot(v[0], k6[0]) + dot(v[1], k6[1]);
149
        sum[1].a += dot(v[0], k7[0]) + dot(v[1], k7[1]);
150
    }
151
#endif
152

153
    sum = activation_afpvec8(sum, activation_type, activation_param_0, activation_param_1);
154

155
#if NCNN_image_shader
156
    image3d_st8(top_blob, ivec3(gx, 0, 0), sum);
157
#else
158
    buffer_st8(top_blob_data, gx, sum);
159
#endif
160
}
161