1/*
2 * Copyright (c) 2015 Advanced Micro Devices, Inc.
3 *
4 * Permission is hereby granted, free of charge, to any person obtaining a copy
5 * of this software and associated documentation files (the "Software"), to deal
6 * in the Software without restriction, including without limitation the rights
7 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
8 * copies of the Software, and to permit persons to whom the Software is
9 * furnished to do so, subject to the following conditions:
10 *
11 * The above copyright notice and this permission notice shall be included in
12 * all copies or substantial portions of the Software.
13 *
14 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
15 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
16 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
17 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
18 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
19 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
20 * THE SOFTWARE.
21 */
22
23#include <clc/clc.h>
24#include "../../../generic/lib/clcmacro.h"
25#include "math/clc_sqrt.h"
26
27_CLC_DEFINE_UNARY_BUILTIN(float, sqrt, __clc_sqrt, float)
28
29#ifdef cl_khr_fp64
30
31#pragma OPENCL EXTENSION cl_khr_fp64 : enable
32
33
34_CLC_OVERLOAD _CLC_DEF double sqrt(double x) {
35
36  uint vcc = x < 0x1p-767;
37  uint exp0 = vcc ? 0x100 : 0;
38  unsigned exp1 = vcc ? 0xffffff80 : 0;
39
40  double v01 = ldexp(x, exp0);
41  double v23 = __builtin_amdgpu_rsq(v01);
42  double v45 = v01 * v23;
43  v23 = v23 * 0.5;
44
45  double v67 = fma(-v23, v45, 0.5);
46  v45 = fma(v45, v67, v45);
47  double v89 = fma(-v45, v45, v01);
48  v23 = fma(v23, v67, v23);
49  v45 = fma(v89, v23, v45);
50  v67 = fma(-v45, v45, v01);
51  v23 = fma(v67, v23, v45);
52
53  v23 = ldexp(v23, exp1);
54  return ((x == __builtin_inf()) || (x == 0.0)) ? v01 : v23;
55}
56
57_CLC_UNARY_VECTORIZE(_CLC_OVERLOAD _CLC_DEF, double, sqrt, double);
58
59#endif
60