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
|
#include "../../prelude/slang-cpp-prelude.h"
namespace { // anonymous
#ifdef SLANG_PRELUDE_NAMESPACE
using namespace SLANG_PRELUDE_NAMESPACE;
#endif
struct KernelContext;
#line 13 "shader.slang"
struct UniformState
{
#line 4
RWStructuredBuffer<float> ioBuffer_0;
};
struct KernelContext
{
UniformState* uniformState;
uint3 dispatchThreadID;
uint3 groupID;
uint3 groupDispatchThreadID;
uint3 calcGroupThreadID() const
{
uint3 v = { dispatchThreadID.x - groupDispatchThreadID.x, dispatchThreadID.y - groupDispatchThreadID.y, dispatchThreadID.z - groupDispatchThreadID.z };
return v;
}
#line 8
void _computeMain()
{
#line 10
uint32_t tid_0 = dispatchThreadID.x;
float i_0 = (uniformState->ioBuffer_0)[tid_0];
bool _S1 = i_0 < 0.50000000000000000000f;
#line 13
float _S2 = i_0 + i_0;
#line 13
float _S3 = (F32_sqrt((i_0)));
#line 13
float o_0 = _S1 ? _S2 : _S3;
(uniformState->ioBuffer_0)[tid_0] = o_0;
#line 8
return;
}
};
} // anonymous
// [numthreads(4, 1, 1)]
SLANG_PRELUDE_EXPORT
void computeMain_Thread(ComputeThreadVaryingInput* varyingInput, void* params, void* uniformState)
{
KernelContext context = {};
context.uniformState = (UniformState*)uniformState;
context.dispatchThreadID = {
varyingInput->groupID.x * 4 + varyingInput->groupThreadID.x,
varyingInput->groupID.y * 1 + varyingInput->groupThreadID.y,
varyingInput->groupID.z * 1 + varyingInput->groupThreadID.z
};
context._computeMain();
}
// [numthreads(4, 1, 1)]
SLANG_PRELUDE_EXPORT
void computeMain_Group(ComputeVaryingInput* varyingInput, void* params, void* uniformState)
{
KernelContext context = {};
context.uniformState = (UniformState*)uniformState;
const uint3 start = {
varyingInput->startGroupID.x * 4,
varyingInput->startGroupID.y * 1,
varyingInput->startGroupID.z * 1
};
context.dispatchThreadID = start;
for (uint32_t x = start.x; x < start.x + 4; ++x)
{
context.dispatchThreadID.x = x;
context._computeMain();
}
}
// [numthreads(4, 1, 1)]
SLANG_PRELUDE_EXPORT
void computeMain(ComputeVaryingInput* varyingInput, void* params, void* uniformState)
{
KernelContext context = {};
context.uniformState = (UniformState*)uniformState;
const uint3 start = {
varyingInput->startGroupID.x * 4,
varyingInput->startGroupID.y * 1,
varyingInput->startGroupID.z * 1
};
const uint3 end = {
varyingInput->endGroupID.x * 4,
varyingInput->endGroupID.y * 1,
varyingInput->endGroupID.z * 1
};
for (uint32_t z = start.z; z < end.z; ++z)
{
context.dispatchThreadID.z = z;
for (uint32_t y = start.y; y < end.y; ++y)
{
context.dispatchThreadID.y = y;
for (uint32_t x = start.x; x < end.x; ++x)
{
context.dispatchThreadID.x = x;
context._computeMain();
}
}
}
}
|
