blob: 22b2add3d37f69cc10bbeba450adec88f37236cc (
plain)
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
|
// atomics-groupshared.slang
//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -shaderobj
//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12 -shaderobj
//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -vk -shaderobj
//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -cuda -shaderobj
//TEST(compute):COMPARE_COMPUTE:-slang -shaderobj -mtl
// Not supported in WGSL: Use of traditional atomics intrinsics (InterlockedXXX functions)
//DISABLE_TEST(compute):COMPARE_COMPUTE:-slang -shaderobj -wgpu
//TEST_INPUT:ubuffer(data=[0 0 0 0], stride=4):out,name outputBuffer
RWStructuredBuffer<uint> outputBuffer;
groupshared uint shared[4];
uint test(uint val)
{
uint originalValue;
shared[val] = 0;
GroupMemoryBarrierWithGroupSync();
uint originalSum = 0;
InterlockedAdd(shared[val], val, originalValue);
originalSum += originalValue;
InterlockedAdd(shared[val ^ 1], val*16, originalValue);
originalSum += originalValue;
InterlockedAdd(shared[val ^ 2], val*16*16, originalValue);
originalSum += originalValue;
GroupMemoryBarrierWithGroupSync();
return shared[val] ^ originalSum;
}
[numthreads(4, 1, 1)]
void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
{
uint tid = dispatchThreadID.x;
uint val = test(tid);
outputBuffer[tid] = val;
}
|
