95 lines
2.1 KiB
Plaintext
95 lines
2.1 KiB
Plaintext
#pragma clang diagnostic ignored "-Wmissing-prototypes"
|
|
|
|
#include <metal_stdlib>
|
|
#include <simd/simd.h>
|
|
|
|
using namespace metal;
|
|
|
|
struct Data
|
|
{
|
|
float a;
|
|
float b;
|
|
};
|
|
|
|
constant float X_tmp [[function_constant(0)]];
|
|
constant float X = is_function_constant_defined(X_tmp) ? X_tmp : 4.0;
|
|
|
|
struct Data_1
|
|
{
|
|
float a;
|
|
float b;
|
|
};
|
|
|
|
struct SSBO
|
|
{
|
|
Data_1 outdata[1];
|
|
};
|
|
|
|
constant uint3 gl_WorkGroupSize [[maybe_unused]] = uint3(2u, 1u, 1u);
|
|
|
|
constant Data _25[2] = { Data{ 1.0, 2.0 }, Data{ 3.0, 4.0 } };
|
|
|
|
template<typename T, uint A>
|
|
void spvArrayCopyFromConstantToStack1(thread T (&dst)[A], constant T (&src)[A])
|
|
{
|
|
for (uint i = 0; i < A; i++)
|
|
{
|
|
dst[i] = src[i];
|
|
}
|
|
}
|
|
|
|
template<typename T, uint A>
|
|
void spvArrayCopyFromConstantToThreadGroup1(threadgroup T (&dst)[A], constant T (&src)[A])
|
|
{
|
|
for (uint i = 0; i < A; i++)
|
|
{
|
|
dst[i] = src[i];
|
|
}
|
|
}
|
|
|
|
template<typename T, uint A>
|
|
void spvArrayCopyFromStackToStack1(thread T (&dst)[A], thread const T (&src)[A])
|
|
{
|
|
for (uint i = 0; i < A; i++)
|
|
{
|
|
dst[i] = src[i];
|
|
}
|
|
}
|
|
|
|
template<typename T, uint A>
|
|
void spvArrayCopyFromStackToThreadGroup1(threadgroup T (&dst)[A], thread const T (&src)[A])
|
|
{
|
|
for (uint i = 0; i < A; i++)
|
|
{
|
|
dst[i] = src[i];
|
|
}
|
|
}
|
|
|
|
template<typename T, uint A>
|
|
void spvArrayCopyFromThreadGroupToStack1(thread T (&dst)[A], threadgroup const T (&src)[A])
|
|
{
|
|
for (uint i = 0; i < A; i++)
|
|
{
|
|
dst[i] = src[i];
|
|
}
|
|
}
|
|
|
|
template<typename T, uint A>
|
|
void spvArrayCopyFromThreadGroupToThreadGroup1(threadgroup T (&dst)[A], threadgroup const T (&src)[A])
|
|
{
|
|
for (uint i = 0; i < A; i++)
|
|
{
|
|
dst[i] = src[i];
|
|
}
|
|
}
|
|
|
|
kernel void main0(device SSBO& _53 [[buffer(0)]], uint3 gl_WorkGroupID [[threadgroup_position_in_grid]], uint3 gl_LocalInvocationID [[thread_position_in_threadgroup]])
|
|
{
|
|
Data _31[2] = { Data{ X, 2.0 }, Data{ 3.0, 5.0 } };
|
|
Data data2[2];
|
|
spvArrayCopyFromStackToStack1(data2, _31);
|
|
_53.outdata[gl_WorkGroupID.x].a = _25[gl_LocalInvocationID.x].a + data2[gl_LocalInvocationID.x].a;
|
|
_53.outdata[gl_WorkGroupID.x].b = _25[gl_LocalInvocationID.x].b + data2[gl_LocalInvocationID.x].b;
|
|
}
|
|
|