static const uint3 gl_WorkGroupSize = uint3(4u, 1u, 1u);

ByteAddressBuffer _22 : register(u0);
RWByteAddressBuffer _44 : register(u1);

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

groupshared float sShared[4];

void comp_main()
{
    sShared[gl_LocalInvocationIndex] = asfloat(_22.Load(gl_GlobalInvocationID.x * 4 + 0));
    GroupMemoryBarrierWithGroupSync();
    _44.Store(gl_GlobalInvocationID.x * 4 + 0, asuint(sShared[(4u - gl_LocalInvocationIndex) - 1u]));
}

[numthreads(4, 1, 1)]
void main(SPIRV_Cross_Input stage_input)
{
    gl_GlobalInvocationID = stage_input.gl_GlobalInvocationID;
    gl_LocalInvocationIndex = stage_input.gl_LocalInvocationIndex;
    comp_main();
}