#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 ElementRef
{
    uint offset;
};

struct ElementTag
{
    uint tag;
    uint flags;
};

struct DrawMonoid
{
    uint path_ix;
    uint clip_ix;
};

struct SceneBuf
{
    uint scene[1];
};

struct DrawMonoid_1
{
    uint path_ix;
    uint clip_ix;
};

struct ParentBuf
{
    DrawMonoid_1 parent[1];
};

struct Alloc
{
    uint offset;
};

struct Config
{
    uint n_elements;
    uint n_pathseg;
    uint width_in_tiles;
    uint height_in_tiles;
    Alloc tile_alloc;
    Alloc bin_alloc;
    Alloc ptcl_alloc;
    Alloc pathseg_alloc;
    Alloc anno_alloc;
    Alloc trans_alloc;
    Alloc bbox_alloc;
    Alloc drawmonoid_alloc;
    uint n_trans;
    uint trans_offset;
    uint pathtag_offset;
    uint linewidth_offset;
    uint pathseg_offset;
};

struct ConfigBuf
{
    Config conf;
};

struct Memory
{
    uint mem_offset;
    uint mem_error;
    uint memory[1];
};

constant uint3 gl_WorkGroupSize [[maybe_unused]] = uint3(512u, 1u, 1u);

static inline __attribute__((always_inline))
ElementTag Element_tag(thread const ElementRef& ref, const device SceneBuf& v_49)
{
    uint tag_and_flags = v_49.scene[ref.offset >> uint(2)];
    return ElementTag{ tag_and_flags & 65535u, tag_and_flags >> uint(16) };
}

static inline __attribute__((always_inline))
DrawMonoid map_tag(thread const uint& tag_word)
{
    switch (tag_word)
    {
        case 4u:
        case 5u:
        case 6u:
        {
            return DrawMonoid{ 1u, 0u };
        }
        case 9u:
        {
            return DrawMonoid{ 1u, 1u };
        }
        case 10u:
        {
            return DrawMonoid{ 0u, 1u };
        }
        default:
        {
            return DrawMonoid{ 0u, 0u };
        }
    }
}

static inline __attribute__((always_inline))
ElementRef Element_index(thread const ElementRef& ref, thread const uint& index)
{
    return ElementRef{ ref.offset + (index * 36u) };
}

static inline __attribute__((always_inline))
DrawMonoid combine_tag_monoid(thread const DrawMonoid& a, thread const DrawMonoid& b)
{
    DrawMonoid c;
    c.path_ix = a.path_ix + b.path_ix;
    c.clip_ix = a.clip_ix + b.clip_ix;
    return c;
}

static inline __attribute__((always_inline))
DrawMonoid tag_monoid_identity()
{
    return DrawMonoid{ 0u, 0u };
}

kernel void main0(device Memory& _277 [[buffer(0)]], const device ConfigBuf& _248 [[buffer(1)]], const device SceneBuf& v_49 [[buffer(2)]], const device ParentBuf& _218 [[buffer(3)]], uint3 gl_GlobalInvocationID [[thread_position_in_grid]], uint3 gl_LocalInvocationID [[thread_position_in_threadgroup]], uint3 gl_WorkGroupID [[threadgroup_position_in_grid]])
{
    threadgroup DrawMonoid sh_scratch[512];
    uint ix = gl_GlobalInvocationID.x * 8u;
    ElementRef ref = ElementRef{ ix * 36u };
    ElementRef param = ref;
    uint tag_word = Element_tag(param, v_49).tag;
    uint param_1 = tag_word;
    DrawMonoid agg = map_tag(param_1);
    spvUnsafeArray<DrawMonoid, 8> local;
    local[0] = agg;
    for (uint i = 1u; i < 8u; i++)
    {
        ElementRef param_2 = ref;
        uint param_3 = i;
        ElementRef param_4 = Element_index(param_2, param_3);
        tag_word = Element_tag(param_4, v_49).tag;
        uint param_5 = tag_word;
        DrawMonoid param_6 = agg;
        DrawMonoid param_7 = map_tag(param_5);
        agg = combine_tag_monoid(param_6, param_7);
        local[i] = agg;
    }
    sh_scratch[gl_LocalInvocationID.x] = agg;
    for (uint i_1 = 0u; i_1 < 9u; i_1++)
    {
        threadgroup_barrier(mem_flags::mem_threadgroup);
        if (gl_LocalInvocationID.x >= (1u << i_1))
        {
            DrawMonoid other = sh_scratch[gl_LocalInvocationID.x - (1u << i_1)];
            DrawMonoid param_8 = other;
            DrawMonoid param_9 = agg;
            agg = combine_tag_monoid(param_8, param_9);
        }
        threadgroup_barrier(mem_flags::mem_threadgroup);
        sh_scratch[gl_LocalInvocationID.x] = agg;
    }
    threadgroup_barrier(mem_flags::mem_threadgroup);
    DrawMonoid row = tag_monoid_identity();
    if (gl_WorkGroupID.x > 0u)
    {
        uint _221 = gl_WorkGroupID.x - 1u;
        row.path_ix = _218.parent[_221].path_ix;
        row.clip_ix = _218.parent[_221].clip_ix;
    }
    if (gl_LocalInvocationID.x > 0u)
    {
        DrawMonoid param_10 = row;
        DrawMonoid param_11 = sh_scratch[gl_LocalInvocationID.x - 1u];
        row = combine_tag_monoid(param_10, param_11);
    }
    uint out_base = (_248.conf.drawmonoid_alloc.offset >> uint(2)) + ((gl_GlobalInvocationID.x * 2u) * 8u);
    for (uint i_2 = 0u; i_2 < 8u; i_2++)
    {
        DrawMonoid param_12 = row;
        DrawMonoid param_13 = local[i_2];
        DrawMonoid m = combine_tag_monoid(param_12, param_13);
        _277.memory[out_base + (i_2 * 2u)] = m.path_ix;
        _277.memory[(out_base + (i_2 * 2u)) + 1u] = m.clip_ix;
    }
}