#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];
    }
};

struct SSBO
{
    spvUnsafeArray<float4, 1> in_data;
};

struct SSBO2
{
    spvUnsafeArray<float4, 1> out_data;
};

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

// Implementation of the GLSL mod() function, which is slightly different than Metal fmod()
template<typename Tx, typename Ty>
inline Tx mod(Tx x, Ty y)
{
    return x - y * floor(x / y);
}

kernel void main0(const device SSBO& _23 [[buffer(0)]], device SSBO2& _33 [[buffer(1)]], uint3 gl_GlobalInvocationID [[thread_position_in_grid]])
{
    uint ident = gl_GlobalInvocationID.x;
    float4 v = mod(_23.in_data[ident], _33.out_data[ident]);
    _33.out_data[ident] = v;
    uint4 vu = as_type<uint4>(_23.in_data[ident]) % as_type<uint4>(_33.out_data[ident]);
    _33.out_data[ident] = as_type<float4>(vu);
    int4 vi = as_type<int4>(_23.in_data[ident]) % as_type<int4>(_33.out_data[ident]);
    _33.out_data[ident] = as_type<float4>(vi);
}