74 lines
2.1 KiB
Plaintext
74 lines
2.1 KiB
Plaintext
#pragma clang diagnostic ignored "-Wmissing-prototypes"
|
|
#pragma clang diagnostic ignored "-Wmissing-braces"
|
|
#pragma clang diagnostic ignored "-Wunused-variable"
|
|
|
|
#include <metal_stdlib>
|
|
#include <simd/simd.h>
|
|
|
|
using namespace metal;
|
|
|
|
template<typename T, size_t Num>
|
|
struct spvUnsafeArray
|
|
{
|
|
T elements[Num ? Num : 1];
|
|
|
|
thread T& operator [] (size_t pos) thread
|
|
{
|
|
return elements[pos];
|
|
}
|
|
constexpr const thread T& operator [] (size_t pos) const thread
|
|
{
|
|
return elements[pos];
|
|
}
|
|
|
|
device T& operator [] (size_t pos) device
|
|
{
|
|
return elements[pos];
|
|
}
|
|
constexpr const device T& operator [] (size_t pos) const device
|
|
{
|
|
return elements[pos];
|
|
}
|
|
|
|
constexpr const constant T& operator [] (size_t pos) const constant
|
|
{
|
|
return elements[pos];
|
|
}
|
|
|
|
threadgroup T& operator [] (size_t pos) threadgroup
|
|
{
|
|
return elements[pos];
|
|
}
|
|
constexpr const threadgroup T& operator [] (size_t pos) const threadgroup
|
|
{
|
|
return elements[pos];
|
|
}
|
|
};
|
|
|
|
struct SSBO
|
|
{
|
|
spvUnsafeArray<float, 1> out_data;
|
|
};
|
|
|
|
constant uint3 gl_WorkGroupSize [[maybe_unused]] = uint3(4u, 4u, 1u);
|
|
|
|
static inline __attribute__((always_inline))
|
|
void work(threadgroup float (&foo)[4][4], thread uint3& gl_LocalInvocationID, thread uint& gl_LocalInvocationIndex, device SSBO& v_67, thread uint3& gl_GlobalInvocationID)
|
|
{
|
|
foo[gl_LocalInvocationID.x][gl_LocalInvocationID.y] = float(gl_LocalInvocationIndex);
|
|
threadgroup_barrier(mem_flags::mem_threadgroup);
|
|
float x = 0.0;
|
|
x += foo[gl_LocalInvocationID.x][0];
|
|
x += foo[gl_LocalInvocationID.x][1];
|
|
x += foo[gl_LocalInvocationID.x][2];
|
|
x += foo[gl_LocalInvocationID.x][3];
|
|
v_67.out_data[gl_GlobalInvocationID.x] = x;
|
|
}
|
|
|
|
kernel void main0(device SSBO& v_67 [[buffer(0)]], uint3 gl_LocalInvocationID [[thread_position_in_threadgroup]], uint gl_LocalInvocationIndex [[thread_index_in_threadgroup]], uint3 gl_GlobalInvocationID [[thread_position_in_grid]])
|
|
{
|
|
threadgroup float foo[4][4];
|
|
work(foo, gl_LocalInvocationID, gl_LocalInvocationIndex, v_67, gl_GlobalInvocationID);
|
|
}
|
|
|