about summary refs log tree commit diff
path: root/sysdeps/libm-ieee754/e_exp.c
blob: 660c5bc88db5fe51e5e58d0c331c7be09ad401de (plain) (blame)
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
171
172
173
174
175
176
177
178
/* Double-precision floating point e^x.
   Copyright (C) 1997, 1998 Free Software Foundation, Inc.
   This file is part of the GNU C Library.
   Contributed by Geoffrey Keating <geoffk@ozemail.com.au>

   The GNU C Library is free software; you can redistribute it and/or
   modify it under the terms of the GNU Library General Public License as
   published by the Free Software Foundation; either version 2 of the
   License, or (at your option) any later version.

   The GNU C Library is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
   Library General Public License for more details.

   You should have received a copy of the GNU Library General Public
   License along with the GNU C Library; see the file COPYING.LIB.  If not,
   write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
   Boston, MA 02111-1307, USA.  */

/* How this works:
   The basic design here is from
   Shmuel Gal and Boris Bachelis, "An Accurate Elementary Mathematical
   Library for the IEEE Floating Point Standard", ACM Trans. Math. Soft.,
   17 (1), March 1991, pp. 26-45.

   The input value, x, is written as

   x = n * ln(2)_0 + t/512 + delta[t] + x + n * ln(2)_1

   where:
   - n is an integer, 1024 >= n >= -1075;
   - ln(2)_0 is the first 43 bits of ln(2), and ln(2)_1 is the remainder, so
     that |ln(2)_1| < 2^-32;
   - t is an integer, 177 >= t >= -177
   - delta is based on a table entry, delta[t] < 2^-28
   - x is whatever is left, |x| < 2^-10

   Then e^x is approximated as

   e^x = 2^n_1 ( 2^n_0 e^(t/512 + delta[t])
               + ( 2^n_0 e^(t/512 + delta[t])
                   * ( p(x + n * ln(2)_1)
                       - n*ln(2)_1
                       - n*ln(2)_1 * p(x + n * ln(2)_1) ) ) )

   where
   - p(x) is a polynomial approximating e(x)-1;
   - e^(t/512 + delta[t]) is obtained from a table;
   - n_1 + n_0 = n, so that |n_0| < DBL_MIN_EXP-1.

   If it happens that n_1 == 0 (this is the usual case), that multiplication
   is omitted.
   */
#ifndef _GNU_SOURCE
#define _GNU_SOURCE
#endif
#include <float.h>
#include <ieee754.h>
#include <math.h>
#include <fenv.h>
#include <inttypes.h>
#include <math_private.h>

extern const float __exp_deltatable[178];
extern const double __exp_atable[355] /* __attribute__((mode(DF))) */;

static const volatile double TWO1023 = 8.988465674311579539e+307;
static const volatile double TWOM1000 = 9.3326361850321887899e-302;

double
__ieee754_exp (double x)
{
  static const double himark = 709.7827128933840868;
  static const double lomark = -745.1332191019412221;
  /* Check for usual case.  */
  if (isless (x, himark) && isgreater (x, lomark))
    {
      static const double TWO43 = 8796093022208.0;
      static const double TWO52 = 4503599627370496.0;
      /* 1/ln(2).  */
      static const double M_1_LN2 = 1.442695040888963387;
      /* ln(2), part 1 */
      static const double M_LN2_0 = .6931471805598903302;
      /* ln(2), part 2 */
      static const double M_LN2_1 = 5.497923018708371155e-14;

      int tval, unsafe, n_i;
      double x22, n, t, dely, result;
      union ieee754_double ex2_u, scale_u;
      fenv_t oldenv;

      feholdexcept (&oldenv);
      fesetround (FE_TONEAREST);

      /* Calculate n.  */
      if (x >= 0)
	{
	  n = x * M_1_LN2 + TWO52;
	  n -= TWO52;
	}
      else
	{
	  n = x * M_1_LN2 - TWO52;
	  n += TWO52;
	}
      x = x - n*M_LN2_0;
      if (x >= 0)
	{
	  /* Calculate t/512.  */
	  t = x + TWO43;
	  t -= TWO43;
	  x -= t;

	  /* Compute tval = t.  */
	  tval = (int) (t * 512.0);

	  x -= __exp_deltatable[tval];
	}
      else
	{
	  /* As above, but x is negative.  */
	  t = x - TWO43;
	  t += TWO43;
	  x -= t;

	  tval = (int) (t * 512.0);

	  x += __exp_deltatable[-tval];
	}

      /* Now, the variable x contains x + n*ln(2)_1.  */
      dely = n*M_LN2_1;

      /* Compute ex2 = 2^n_0 e^(t/512+delta[t]).  */
      ex2_u.d = __exp_atable[tval+177];
      n_i = (int)n;
      /* 'unsafe' is 1 iff n_1 != 0.  */
      unsafe = abs(n_i) >= -DBL_MIN_EXP - 1;
      ex2_u.ieee.exponent += n_i >> unsafe;

      /* Compute scale = 2^n_1.  */
      scale_u.d = 1.0;
      scale_u.ieee.exponent += n_i - (n_i >> unsafe);

      /* Approximate e^x2 - 1, using a fourth-degree polynomial,
	 with maximum error in [-2^-10-2^-28,2^-10+2^-28]
	 less than 4.9e-19.  */
      x22 = (((0.04166666898464281565
	       * x + 0.1666666766008501610)
	      * x + 0.499999999999990008)
	     * x + 0.9999999999999976685) * x;
      /* Allow for impact of dely.  */
      x22 -= dely + dely*x22;

      /* Return result.  */
      fesetenv (&oldenv);

      result = x22 * ex2_u.d + ex2_u.d;
      if (!unsafe)
	return result;
      else
	return result * scale_u.d;
    }
  /* Exceptional cases:  */
  else if (isless (x, himark))
    {
      if (__isinf (x))
	/* e^-inf == 0, with no error.  */
	return 0;
      else
	/* Underflow */
	return TWOM1000 * TWOM1000;
    }
  else
    /* Return x, if x is a NaN or Inf; or overflow, otherwise.  */
    return TWO1023*x;
}