#version 450

#include "types.comp"
#include "generic_unary_head.comp"

const uint num_threads = 128;

layout(local_size_x = num_threads, local_size_y = 1, local_size_z = 1) in;

void main() {
    uint idx = get_idx();

    // num_threads * num_iter must equal 512, to match the wg_denoms and get_idx calculation
    const uint num_iter = 4;

    [[unroll]] for (uint i = 0; i < num_iter; ++i) {
        if (idx >= p.ne) {
            continue;
        }

        data_d[p.d_offset + idx] = D_TYPE(FLOAT_TYPE(data_a[idx]) * FLOAT_TYPE(p.param1));
        idx += num_threads;
    }
}