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
|
// 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;
//TEST_INPUT: ubuffer(data=[0 0 0 0 0 0 0], stride=4):out,name cudaOutputBuffer
RWStructuredBuffer<float> cudaOutputBuffer;
[numthreads(7, 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;
}
}
|
