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
|
/* -------------------------------------------------------------- */
/* (C)Copyright 2007,2008, */
/* International Business Machines Corporation */
/* All Rights Reserved. */
/* */
/* Redistribution and use in source and binary forms, with or */
/* without modification, are permitted provided that the */
/* following conditions are met: */
/* */
/* - Redistributions of source code must retain the above copyright*/
/* notice, this list of conditions and the following disclaimer. */
/* */
/* - Redistributions in binary form must reproduce the above */
/* copyright notice, this list of conditions and the following */
/* disclaimer in the documentation and/or other materials */
/* provided with the distribution. */
/* */
/* - Neither the name of IBM Corporation nor the names of its */
/* contributors may be used to endorse or promote products */
/* derived from this software without specific prior written */
/* permission. */
/* */
/* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND */
/* CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, */
/* INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF */
/* MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE */
/* DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR */
/* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, */
/* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT */
/* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; */
/* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) */
/* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN */
/* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR */
/* OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, */
/* EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */
/* -------------------------------------------------------------- */
/* PROLOG END TAG zYx */
#ifdef __SPU__
#ifndef _TANHF4_H_
#define _TANHF4_H_ 1
#include <spu_intrinsics.h>
#include "expf4.h"
#include "divf4.h"
#include "tanhd2.h"
/*
* FUNCTION
* vector float _tanhf4(vector float x)
*
* DESCRIPTION
* The _tanhf4 function computes the hyperbolic tangent for each
* element of the input vector.
*
* We use the following to approximate tanh:
*
* |x| <= .25: Taylor Series
* |x| > .25: tanh(x) = (exp(2x) - 1)/(exp(2x) + 1)
*
*
* SPECIAL CASES:
* - tanh(+/- 0) = +/-0
* - tanh(+/- infinity) = +/- 1
*
*/
static __inline vector float _tanhf4(vector float x)
{
vector float signbit = spu_splats(-0.0f);
vector float onef = spu_splats(1.0f);
vector float twof = spu_splats(2.0f);
vector float xabs;
vector float x2;
vector unsigned int gttaylor;
vector float e;
vector float tresult;
vector float eresult;
vector float result;
xabs = spu_andc(x, signbit);
/*
* This is where we switch from Taylor Series
* to exponential formula.
*/
gttaylor = spu_cmpgt(xabs, spu_splats(0.25f));
/*
* Taylor Series Approximation
*/
x2 = spu_mul(x,x);
tresult = spu_madd(x2, spu_splats((float)TANH_TAY06), spu_splats((float)TANH_TAY05));
tresult = spu_madd(x2, tresult, spu_splats((float)TANH_TAY04));
tresult = spu_madd(x2, tresult, spu_splats((float)TANH_TAY03));
tresult = spu_madd(x2, tresult, spu_splats((float)TANH_TAY02));
tresult = spu_madd(x2, tresult, spu_splats((float)TANH_TAY01));
tresult = spu_mul(xabs, tresult);
/*
* Exponential Formula
* Our expf4 function gives a more accurate result in general
* with xabs instead of x for x<0. We correct for sign later.
*/
e = _expf4(spu_mul(xabs, twof));
eresult = _divf4(spu_sub(e, onef), spu_add(e, onef));
/*
* Select Taylor or exp result.
*/
result = spu_sel(tresult, eresult, gttaylor);
/*
* Correct for accumulated truncation error when
* tanh(x) should return 1.
* Note that this also handles the special case of
* x = +/- infinity.
*/
result = spu_sel(result, onef, spu_cmpgt(xabs, spu_splats(9.125f)));
/*
* Antisymmetric function - preserve sign bit of x
* in the result.
*/
result = spu_sel(result, x, (vec_uint4)signbit);
return result;
}
#endif /* _TANHF4_H_ */
#endif /* __SPU__ */
|
