/*
  Copyright (c) 2021-2024, Intel Corporation

  SPDX-License-Identifier: BSD-3-Clause
*/
#ifndef ISPC_GPU
#if (TEST_SIG == 0)
task void f_v(uniform float res[]);
export void f_v_cpu_entry_point(uniform float res[]) { launch[1] f_v(res); }
#elif (TEST_SIG == 1)
task void f_f(uniform float res[], uniform float vfloat[]);
export void f_f_cpu_entry_point(uniform float res[], uniform float vfloat[]) { launch[1] f_f(res, vfloat); }
#elif (TEST_SIG == 2)
task void f_fu(uniform float res[], uniform float vfloat[], uniform float b);
export void f_fu_cpu_entry_point(uniform float res[], uniform float vfloat[], uniform float b) {
    launch[1] f_fu(res, vfloat, b);
}
#elif (TEST_SIG == 3)
task void f_fi(uniform float res[], uniform float vfloat[], uniform int vint[]);
export void f_fi_cpu_entry_point(uniform float res[], uniform float vfloat[], uniform int vint[]) {
    launch[1] f_fi(res, vfloat, vint);
}
#elif (TEST_SIG == 4)
task void f_du(uniform float res[], uniform double a[], uniform double b);
export void f_du_cpu_entry_point(uniform float res[], uniform double a[], uniform double b) {
    launch[1] f_du(res, a, b);
}
#elif (TEST_SIG == 5)
task void f_duf(uniform float res[], uniform double a[], uniform float b);
export void f_duf_cpu_entry_point(uniform float res[], uniform double a[], uniform float b) {
    launch[1] f_duf(res, a, b);
}
#elif (TEST_SIG == 6)
task void f_di(uniform float res[], uniform double a[], uniform int b[]);
export void f_di_cpu_entry_point(uniform float res[], uniform double a[], uniform int b[]) {
    launch[1] f_di(res, a, b);
}
#elif (TEST_SIG == 32)
task void print_uf(uniform float b);
export void print_uf_cpu_entry_point(uniform float b) { launch[1] print_uf(b); }
#elif (TEST_SIG == 33)
task void print_f(uniform float vfloat[]);
export void print_f_cpu_entry_point(uniform float vfloat[]) { launch[1] print_f(vfloat); }
#elif (TEST_SIG == 34)
task void print_fuf(uniform float vfloat[], uniform float b);
export void print_fuf_cpu_entry_point(uniform float vfloat[], uniform float b) { launch[1] print_fuf(vfloat, b); }
#endif

#if (TEST_SIG < 32)
task void result(uniform float res[]);
export void result_cpu_entry_point(uniform float res[]) { launch[1] result(res); }
#else
task void print_result();
export void print_result_cpu_entry_point() { launch[1] print_result(); }
#endif
#endif

template <typename T>
void pack4toint(uniform T * uniform input, uniform uint32 * uniform packedOutput, uniform int32 inputSize) {
    for (uniform int i = 0; i < inputSize; i += 4) {
        uniform uint32 packedValue = 0;
        packedValue = ((uniform int32)input[i] << 24) |
                      ((uniform int32)(uniform uint8)input[i+1] << 16) |
                      ((uniform int32)(uniform uint8)input[i+2] << 8) |
                      (uniform int32)(uniform uint8)input[i+3];
        packedOutput[i / 4] = packedValue;
    }
}

template <typename T>
void pack2toint(uniform T * uniform input, uniform uint32 * uniform packedOutput, uniform int32 inputSize) {
    for (uniform int i = 0; i < inputSize; i += 2) {
        uniform uint32 packedValue = 0;
        packedValue = ((uniform int32)input[i] << 16) |
                      (uniform int32)(uniform uint16)input[i+1];
        packedOutput[i / 2] = packedValue;
    }
}