shaders-msl/comp/shared.comp

#version 310 es
layout(local_size_x = 4) in;

shared float sShared[gl_WorkGroupSize.x];

layout(std430, binding = 0) readonly buffer SSBO
{
    float in_data[];
};

layout(std430, binding = 1) writeonly buffer SSBO2
{
    float out_data[];
};

void main()
{
    uint ident = gl_GlobalInvocationID.x;
    float idata = in_data[ident];

    sShared[gl_LocalInvocationIndex] = idata;
    memoryBarrierShared();
    barrier();

    out_data[ident] = sShared[gl_WorkGroupSize.x - gl_LocalInvocationIndex - 1u];
}