struct BlockOut
{
    float4 a;
    float4 b;
};

struct BlockOutPrim
{
    float4 a;
    float4 b;
};

struct TaskPayload
{
    float a;
    float b;
    int c;
};

static const uint3 gl_WorkGroupSize = uint3(2u, 3u, 4u);

static uint3 gl_WorkGroupID;
static uint3 gl_GlobalInvocationID;
static uint gl_LocalInvocationIndex;
struct SPIRV_Cross_Input
{
    uint3 gl_WorkGroupID : SV_GroupID;
    uint3 gl_GlobalInvocationID : SV_DispatchThreadID;
    uint gl_LocalInvocationIndex : SV_GroupIndex;
};

struct gl_MeshPerVertexEXT
{
    float4 vOut : TEXCOORD0;
    BlockOut outputs : TEXCOORD2;
    float4 gl_Position : SV_Position;
    float gl_ClipDistance[1] : SV_ClipDistance;
    float2 gl_CullDistance : SV_CullDistance;
};

struct gl_MeshPerPrimitiveEXT
{
    float4 vPrim : TEXCOORD1;
    BlockOutPrim prim_outputs : TEXCOORD4;
    uint gl_PrimitiveID : SV_PrimitiveID;
    uint gl_Layer : SV_RenderTargetArrayIndex;
    uint gl_ViewportIndex : SV_ViewportArrayIndex;
    uint gl_PrimitiveShadingRateEXT : SV_ShadingRate;
    bool gl_CullPrimitiveEXT : SV_CullPrimitive;
};

groupshared float shared_float[16];

void mesh_main(inout gl_MeshPerVertexEXT gl_MeshVerticesEXT[24], inout gl_MeshPerPrimitiveEXT gl_MeshPrimitivesEXT[22], TaskPayload _payload, inout uint3 gl_PrimitiveTriangleIndicesEXT[22])
{
    SetMeshOutputCounts(24u, 22u);
    gl_MeshVerticesEXT[gl_LocalInvocationIndex].gl_Position = float4(float3(gl_GlobalInvocationID), 1.0f);
    gl_MeshVerticesEXT[gl_LocalInvocationIndex].gl_ClipDistance[0] = 4.0f;
    gl_MeshVerticesEXT[gl_LocalInvocationIndex].gl_CullDistance[0] = 3.0f;
    gl_MeshVerticesEXT[gl_LocalInvocationIndex].gl_CullDistance[1] = 5.0f;
    gl_MeshVerticesEXT[gl_LocalInvocationIndex].vOut = float4(float3(gl_GlobalInvocationID), 2.0f);
    gl_MeshVerticesEXT[gl_LocalInvocationIndex].outputs.a = 5.0f.xxxx;
    gl_MeshVerticesEXT[gl_LocalInvocationIndex].outputs.b = 6.0f.xxxx;
    GroupMemoryBarrierWithGroupSync();
    if (gl_LocalInvocationIndex < 22u)
    {
        gl_MeshPrimitivesEXT[gl_LocalInvocationIndex].vPrim = float4(float3(gl_WorkGroupID), 3.0f);
        gl_MeshPrimitivesEXT[gl_LocalInvocationIndex].prim_outputs.a = _payload.a.xxxx;
        gl_MeshPrimitivesEXT[gl_LocalInvocationIndex].prim_outputs.b = _payload.b.xxxx;
        gl_PrimitiveTriangleIndicesEXT[gl_LocalInvocationIndex] = uint3(0u, 1u, 2u) + gl_LocalInvocationIndex.xxx;
        gl_MeshPrimitivesEXT[gl_LocalInvocationIndex].gl_PrimitiveID = int(gl_GlobalInvocationID.x);
        gl_MeshPrimitivesEXT[gl_LocalInvocationIndex].gl_Layer = int(gl_GlobalInvocationID.x) + 1;
        gl_MeshPrimitivesEXT[gl_LocalInvocationIndex].gl_ViewportIndex = int(gl_GlobalInvocationID.x) + 2;
        gl_MeshPrimitivesEXT[gl_LocalInvocationIndex].gl_CullPrimitiveEXT = (gl_GlobalInvocationID.x & 1u) != 0u;
        gl_MeshPrimitivesEXT[gl_LocalInvocationIndex].gl_PrimitiveShadingRateEXT = int(gl_GlobalInvocationID.x) + 3;
    }
}

[outputtopology("triangle")]
[numthreads(2, 3, 4)]
void main(SPIRV_Cross_Input stage_input, out vertices gl_MeshPerVertexEXT gl_MeshVerticesEXT[24], out primitives gl_MeshPerPrimitiveEXT gl_MeshPrimitivesEXT[22], in payload TaskPayload _payload, out indices uint3 gl_PrimitiveTriangleIndicesEXT[22])
{
    gl_WorkGroupID = stage_input.gl_WorkGroupID;
    gl_GlobalInvocationID = stage_input.gl_GlobalInvocationID;
    gl_LocalInvocationIndex = stage_input.gl_LocalInvocationIndex;
    mesh_main(gl_MeshVerticesEXT, gl_MeshPrimitivesEXT, _payload, gl_PrimitiveTriangleIndicesEXT);
}