1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
|
/*
* Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*
*/
#if defined(TARGET_LINUX_POWER)
#include "xmm2altivec.h"
#elif defined(TARGET_LINUX_ARM64)
#include "arm64intrin.h"
#else
#include <immintrin.h>
#endif
#include <math.h>
#include "rslog_defs.h"
#define FMAF __builtin_fmaf
#define FMA __builtin_fma
extern "C" float __rss_log_fma3(float);
#ifndef PRECISION
#define PRECISION 1
#endif
#if PRECISION <= 1
float __rss_log_fma3(float a_input)
{
union U {
float f;
unsigned u;
};
struct float4 {
float a, b, c, d;
};
const static float4* C4 = (const float4*)C_F;
U CANONICAL_NAN, N_INF, P_INF;
CANONICAL_NAN.u = CANONICAL_NAN_I;
N_INF.u = NEG_INF;
P_INF.u = POS_INF;
unsigned const bit_mask2 = BIT_MASK2;
unsigned const offset = OFFSET;
unsigned exp_offset = EXP_OFFSET;
U input_union;
input_union.f = a_input;
if (__builtin_expect(!(TWO_TO_M126_F <= a_input), 0))
{
if (a_input != a_input)
{
return a_input + a_input;
}
return a_input < 0.0f ? CANONICAL_NAN.f : N_INF.f;
}
U mantisa_union;
mantisa_union.u = input_union.u & bit_mask2;
int idx = mantisa_union.u >> 17;
int pred = idx >= 30;
float4 coeff = C4[idx];
if (__builtin_expect(a_input == P_INF.f, 0))
{
return a_input;
}
mantisa_union.u += offset;
float m = mantisa_union.f;
int e_int = (input_union.u >> 23) - exp_offset;
float e = (float)e_int;
e = pred ? e : e - 1.0f;
m = pred ? m - 1.0f : FMAF(m, 2.0f, -1.0f);
float exp = FMAF(e, LN2_F, coeff.d);
float t = FMAF(coeff.a, m, coeff.b);
t = FMAF(t, m, coeff.c);
t = FMAF(t, m, exp);
return t;
}
#endif
#if PRECISION >= 2
float __rss_log_fma3(float a_input)
{
union U {
float f;
unsigned u;
};
struct double4 {
double a, b, c, d;
};
const static double4* C4 = (const double4*)C_D;
U CANONICAL_NAN, N_INF, P_INF;
CANONICAL_NAN.u = CANONICAL_NAN_I;
N_INF.u = NEG_INF;
P_INF.u = POS_INF;
unsigned const bit_mask2 = BIT_MASK2;
unsigned long long const offset = OFFSET;
unsigned const exp_offset = EXP_OFFSET;
U input_union;
input_union.f = a_input;
if (__builtin_expect(!(TWO_TO_M126_F <= a_input), 0))
{
if (a_input != a_input)
{
return a_input;
}
return a_input < 0.0f ? CANONICAL_NAN.f : N_INF.f;
}
U mantisa_union;
mantisa_union.u = input_union.u & bit_mask2;
int idx = mantisa_union.u >> 17;
int pred = mantisa_union.u >> 22;
double4 coeff = C4[idx];
if (__builtin_expect(a_input == P_INF.f, 0))
{
return a_input;
}
mantisa_union.u += offset;
double m = (double)mantisa_union.f;
int e_int = (input_union.u>>23) - exp_offset;
double e = (double)e_int;
e = pred ? e : e - 1.0f;
m = pred ? m - 1.0f : FMAF(m, 2.0f, -1.0f);
double t;
t = FMA(coeff.a, m, coeff.b);
t = FMA(t, m, coeff.c);
t = FMA(t, m, coeff.d);
t = FMA(t, m, e * LN2_D);
return (float)t;
}
#endif
|
