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
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
|
// Test for verifying subscript operator support in cuda.
//TEST(compute):COMPARE_COMPUTE_EX:-cuda -compute -shaderobj
//TEST_INPUT: Texture1D(size=4, content = one):name cudaT1D
Texture1D<float> cudaT1D;
//TEST_INPUT: Texture2D(size=8, content = one):name cudaT2D
Texture2D<float> cudaT2D;
//TEST_INPUT: Texture2D(size=8, content = one):name cudaT2D_f2
Texture2D<float2> cudaT2D_f2;
//TEST_INPUT: Texture2D(size=8, content = one):name cudaT2D_f4
Texture2D<float4> cudaT2D_f4;
//TEST_INPUT: Texture3D(size=8, content = one):name cudaT3D
Texture3D<float> cudaT3D;
//TEST_INPUT: Texture3D(size=8, content = one):name cudaT3D_f2
Texture3D<float2> cudaT3D_f2;
//TEST_INPUT: Texture3D(size=8, content = one):name cudaT3D_f4
Texture3D<float4> cudaT3D_f4;
//TEST_INPUT: TextureCube(size=16, content = one):name cudaTCube
TextureCube<float> cudaTCube;
//TEST_INPUT: Texture2D(size=16, content = one, arrayLength=3):name cudaT2DArray
Texture2DArray<float> cudaT2DArray;
//TEST_INPUT: Texture2D(size=16, content = one, arrayLength=3):name cudaT2DArray_f2
Texture2DArray<float2> cudaT2DArray_f2;
//TEST_INPUT: Texture2D(size=16, content = one, arrayLength=3):name cudaT2DArray_f4
Texture2DArray<float4> cudaT2DArray_f4;
// New texture declarations for uint and int variants
//TEST_INPUT: Texture2D(size=8, content = one):name cudaT2D_uint
Texture2D<uint> cudaT2D_uint;
//TEST_INPUT: Texture2D(size=8, content = one):name cudaT2D_uint2
Texture2D<uint2> cudaT2D_uint2;
//TEST_INPUT: Texture2D(size=8, content = one):name cudaT2D_uint4
Texture2D<uint4> cudaT2D_uint4;
//TEST_INPUT: Texture2D(size=8, content = one):name cudaT2D_int
Texture2D<int> cudaT2D_int;
//TEST_INPUT: Texture2D(size=8, content = one):name cudaT2D_int2
Texture2D<int2> cudaT2D_int2;
//TEST_INPUT: Texture2D(size=8, content = one):name cudaT2D_int4
Texture2D<int4> cudaT2D_int4;
//TEST_INPUT: Texture3D(size=8, content = one):name cudaT3D_uint
Texture3D<uint> cudaT3D_uint;
//TEST_INPUT: Texture3D(size=8, content = one):name cudaT3D_uint2
Texture3D<uint2> cudaT3D_uint2;
//TEST_INPUT: Texture3D(size=8, content = one):name cudaT3D_uint4
Texture3D<uint4> cudaT3D_uint4;
//TEST_INPUT: Texture3D(size=8, content = one):name cudaT3D_int
Texture3D<int> cudaT3D_int;
//TEST_INPUT: Texture3D(size=8, content = one):name cudaT3D_int2
Texture3D<int2> cudaT3D_int2;
//TEST_INPUT: Texture3D(size=8, content = one):name cudaT3D_int4
Texture3D<int4> cudaT3D_int4;
//TEST_INPUT: ubuffer(data=[0 0 0 0 0 0 0 0 0 0 0], stride=4):out,name cudaOutputBuffer
RWStructuredBuffer<float> cudaOutputBuffer;
[numthreads(11, 1, 1)]
[shader("compute")]
void computeMain(int3 dispatchThreadID : SV_DispatchThreadID)
{
int idx = dispatchThreadID.x;
switch (idx)
{
case 1:
{
int var = 0;
float result = cudaT1D[0];
// This is not supported in PTX.
//cudaOutputBuffer[idx] = result;
}
break;
case 2:
{
int2 var = int2(1, 2);
float result = cudaT2D[var];
cudaOutputBuffer[idx] = result;
float2 result2 = cudaT2D_f2[var];
cudaOutputBuffer[idx] += result2.x;
cudaOutputBuffer[idx] += result2.y;
float4 result4 = cudaT2D_f4[var];
cudaOutputBuffer[idx] += result4.x;
cudaOutputBuffer[idx] += result4.y;
cudaOutputBuffer[idx] += result4.z;
cudaOutputBuffer[idx] += result4.w;
}
break;
case 3:
{
int3 var = int3(1, 1, 1);
float result = cudaT3D[var];
cudaOutputBuffer[idx] = result;
float2 result2 = cudaT3D_f2[var];
cudaOutputBuffer[idx] += result2.x;
cudaOutputBuffer[idx] += result2.y;
float4 result4 = cudaT3D_f4[var];
cudaOutputBuffer[idx] += result4.x;
cudaOutputBuffer[idx] += result4.y;
cudaOutputBuffer[idx] += result4.z;
cudaOutputBuffer[idx] += result4.w;
}
break;
case 4:
{
int3 var = int3(0, 0, 1);
float result = cudaT2DArray[var];
cudaOutputBuffer[idx] = result;
float2 result2 = cudaT2DArray_f2[var];
cudaOutputBuffer[idx] += result2.x;
cudaOutputBuffer[idx] += result2.y;
float4 result4 = cudaT2DArray_f4[var];
cudaOutputBuffer[idx] += result4.x;
cudaOutputBuffer[idx] += result4.y;
cudaOutputBuffer[idx] += result4.z;
cudaOutputBuffer[idx] += result4.w;
}
break;
case 5:
{
// Test 2D textures with uint variants
int2 var = int2(1, 2);
uint result = cudaT2D_uint[var];
cudaOutputBuffer[idx] = float(result);
uint2 result2 = cudaT2D_uint2[var];
cudaOutputBuffer[idx] += float(result2.x);
cudaOutputBuffer[idx] += float(result2.y);
uint4 result4 = cudaT2D_uint4[var];
cudaOutputBuffer[idx] += float(result4.x);
cudaOutputBuffer[idx] += float(result4.y);
cudaOutputBuffer[idx] += float(result4.z);
cudaOutputBuffer[idx] += float(result4.w);
}
break;
case 6:
{
// Test 2D textures with int variants
int2 var = int2(1, 2);
int result = cudaT2D_int[var];
cudaOutputBuffer[idx] = float(result);
int2 result2 = cudaT2D_int2[var];
cudaOutputBuffer[idx] += float(result2.x);
cudaOutputBuffer[idx] += float(result2.y);
int4 result4 = cudaT2D_int4[var];
cudaOutputBuffer[idx] += float(result4.x);
cudaOutputBuffer[idx] += float(result4.y);
cudaOutputBuffer[idx] += float(result4.z);
cudaOutputBuffer[idx] += float(result4.w);
}
break;
case 7:
{
// Test 3D textures with uint variants
int3 var = int3(1, 1, 1);
uint result = cudaT3D_uint[var];
cudaOutputBuffer[idx] = float(result);
uint2 result2 = cudaT3D_uint2[var];
cudaOutputBuffer[idx] += float(result2.x);
cudaOutputBuffer[idx] += float(result2.y);
uint4 result4 = cudaT3D_uint4[var];
cudaOutputBuffer[idx] += float(result4.x);
cudaOutputBuffer[idx] += float(result4.y);
cudaOutputBuffer[idx] += float(result4.z);
cudaOutputBuffer[idx] += float(result4.w);
}
break;
case 8:
{
// Test 3D textures with int variants
int3 var = int3(1, 1, 1);
int result = cudaT3D_int[var];
cudaOutputBuffer[idx] = float(result);
int2 result2 = cudaT3D_int2[var];
cudaOutputBuffer[idx] += float(result2.x);
cudaOutputBuffer[idx] += float(result2.y);
int4 result4 = cudaT3D_int4[var];
cudaOutputBuffer[idx] += float(result4.x);
cudaOutputBuffer[idx] += float(result4.y);
cudaOutputBuffer[idx] += float(result4.z);
cudaOutputBuffer[idx] += float(result4.w);
}
break;
}
}
|
