#pragma clang diagnostic ignored "-Wmissing-prototypes"
#pragma clang diagnostic ignored "-Wmissing-braces"

#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];
    }
};

constant uint3 gl_WorkGroupSize [[maybe_unused]] = uint3(8u, 1u, 1u);

template<typename T, uint A>
inline 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>
inline 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>
inline 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>
inline 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>
inline 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>
inline 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(uint gl_LocalInvocationIndex [[thread_index_in_threadgroup]])
{
    threadgroup float shared_group[8][8];
    threadgroup float shared_group_alt[8][8];
    spvUnsafeArray<float, 8> blob;
    for (int i = 0; i < 8; i++)
    {
        blob[i] = float(i);
    }
    spvArrayCopyFromStackToThreadGroup1(shared_group[gl_LocalInvocationIndex], blob.elements);
    threadgroup_barrier(mem_flags::mem_threadgroup);
    spvUnsafeArray<float, 8> copied_blob;
    spvArrayCopyFromThreadGroupToStack1(copied_blob.elements, shared_group[gl_LocalInvocationIndex ^ 1u]);
    spvArrayCopyFromThreadGroupToThreadGroup1(shared_group_alt[gl_LocalInvocationIndex], shared_group[gl_LocalInvocationIndex]);
}