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
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
|
/*
* ====================================================
* Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.
*
* Developed at SunPro, a Sun Microsystems, Inc. business.
* Permission to use, copy, modify, and distribute this
* software is freely granted, provided that this notice
* is preserved.
* ====================================================
*/
/*
* from: @(#)fdlibm.h 5.1 93/09/24
* $FreeBSD: src/lib/msun/src/math_private.h,v 1.20 2005/11/28 04:58:57 bde Exp $
*/
#ifndef _MATH_PRIVATE_H_
#define _MATH_PRIVATE_H_
#include <sys/types.h>
#include <endian.h>
/*
* The original fdlibm code used statements like:
* n0 = ((*(int*)&one)>>29)^1; * index of high word *
* ix0 = *(n0+(int*)&x); * high word of x *
* ix1 = *((1-n0)+(int*)&x); * low word of x *
* to dig two 32 bit words out of the 64 bit IEEE floating point
* value. That is non-ANSI, and, moreover, the gcc instruction
* scheduler gets it wrong. We instead use the following macros.
* Unlike the original code, we determine the endianness at compile
* time, not at run time; I don't see much benefit to selecting
* endianness at run time.
*/
/*
* A union which permits us to convert between a double and two 32 bit
* ints.
*/
#if (__BYTE_ORDER == __BIG_ENDIAN) || (defined(__arm__) && !defined(__VFP_FP__))
typedef union
{
double value;
struct
{
u_int32_t msw;
u_int32_t lsw;
} parts;
} ieee_double_shape_type;
#endif
#if __BYTE_ORDER == __LITTLE_ENDIAN && !(defined(__arm__) && !defined(__VFP_FP__))
typedef union
{
double value;
struct
{
u_int32_t lsw;
u_int32_t msw;
} parts;
} ieee_double_shape_type;
#endif
/* Get two 32 bit ints from a double. */
#define EXTRACT_WORDS(ix0,ix1,d) \
do { \
ieee_double_shape_type ew_u; \
ew_u.value = (d); \
(ix0) = ew_u.parts.msw; \
(ix1) = ew_u.parts.lsw; \
} while (0)
/* Get the more significant 32 bit int from a double. */
#define GET_HIGH_WORD(i,d) \
do { \
ieee_double_shape_type gh_u; \
gh_u.value = (d); \
(i) = gh_u.parts.msw; \
} while (0)
/* Get the less significant 32 bit int from a double. */
#define GET_LOW_WORD(i,d) \
do { \
ieee_double_shape_type gl_u; \
gl_u.value = (d); \
(i) = gl_u.parts.lsw; \
} while (0)
/* Set a double from two 32 bit ints. */
#define INSERT_WORDS(d,ix0,ix1) \
do { \
ieee_double_shape_type iw_u; \
iw_u.parts.msw = (ix0); \
iw_u.parts.lsw = (ix1); \
(d) = iw_u.value; \
} while (0)
/* Set the more significant 32 bits of a double from an int. */
#define SET_HIGH_WORD(d,v) \
do { \
ieee_double_shape_type sh_u; \
sh_u.value = (d); \
sh_u.parts.msw = (v); \
(d) = sh_u.value; \
} while (0)
/* Set the less significant 32 bits of a double from an int. */
#define SET_LOW_WORD(d,v) \
do { \
ieee_double_shape_type sl_u; \
sl_u.value = (d); \
sl_u.parts.lsw = (v); \
(d) = sl_u.value; \
} while (0)
/*
* A union which permits us to convert between a float and a 32 bit
* int.
*/
typedef union
{
float value;
/* FIXME: Assumes 32 bit int. */
unsigned int word;
} ieee_float_shape_type;
/* Get a 32 bit int from a float. */
#define GET_FLOAT_WORD(i,d) \
do { \
ieee_float_shape_type gf_u; \
gf_u.value = (d); \
(i) = gf_u.word; \
} while (0)
/* Set a float from a 32 bit int. */
#define SET_FLOAT_WORD(d,i) \
do { \
ieee_float_shape_type sf_u; \
sf_u.word = (i); \
(d) = sf_u.value; \
} while (0)
#ifdef _COMPLEX_H
/*
* Inline functions that can be used to construct complex values.
*
* The C99 standard intends x+I*y to be used for this, but x+I*y is
* currently unusable in general since gcc introduces many overflow,
* underflow, sign and efficiency bugs by rewriting I*y as
* (0.0+I)*(y+0.0*I) and laboriously computing the full complex product.
* In particular, I*Inf is corrupted to NaN+I*Inf, and I*-0 is corrupted
* to -0.0+I*0.0.
*/
static __inline float complex
cpackf(float x, float y)
{
float complex z;
__real__ z = x;
__imag__ z = y;
return (z);
}
static __inline double complex
cpack(double x, double y)
{
double complex z;
__real__ z = x;
__imag__ z = y;
return (z);
}
static __inline long double complex
cpackl(long double x, long double y)
{
long double complex z;
__real__ z = x;
__imag__ z = y;
return (z);
}
#endif /* _COMPLEX_H */
/*
* ieee style elementary functions
*
* We rename functions here to improve other sources' diffability
* against fdlibm.
*/
#define __ieee754_sqrt sqrt
#define __ieee754_acos acos
#define __ieee754_acosh acosh
#define __ieee754_log log
#define __ieee754_atanh atanh
#define __ieee754_asin asin
#define __ieee754_atan2 atan2
#define __ieee754_exp exp
#define __ieee754_cosh cosh
#define __ieee754_fmod fmod
#define __ieee754_pow pow
#define __ieee754_lgamma lgamma
#define __ieee754_gamma gamma
#define __ieee754_lgamma_r lgamma_r
#define __ieee754_gamma_r gamma_r
#define __ieee754_log10 log10
#define __ieee754_sinh sinh
#define __ieee754_hypot hypot
#define __ieee754_j0 j0
#define __ieee754_j1 j1
#define __ieee754_y0 y0
#define __ieee754_y1 y1
#define __ieee754_jn jn
#define __ieee754_yn yn
#define __ieee754_remainder remainder
#define __ieee754_scalb scalb
#define __ieee754_sqrtf sqrtf
#define __ieee754_acosf acosf
#define __ieee754_acoshf acoshf
#define __ieee754_logf logf
#define __ieee754_atanhf atanhf
#define __ieee754_asinf asinf
#define __ieee754_atan2f atan2f
#define __ieee754_expf expf
#define __ieee754_coshf coshf
#define __ieee754_fmodf fmodf
#define __ieee754_powf powf
#define __ieee754_lgammaf lgammaf
#define __ieee754_gammaf gammaf
#define __ieee754_lgammaf_r lgammaf_r
#define __ieee754_gammaf_r gammaf_r
#define __ieee754_log10f log10f
#define __ieee754_sinhf sinhf
#define __ieee754_hypotf hypotf
#define __ieee754_j0f j0f
#define __ieee754_j1f j1f
#define __ieee754_y0f y0f
#define __ieee754_y1f y1f
#define __ieee754_jnf jnf
#define __ieee754_ynf ynf
#define __ieee754_remainderf remainderf
#define __ieee754_scalbf scalbf
#define __ieee754_ldexpf ldexpf
/* fdlibm kernel function */
#if defined(__ARM_NEON__)
double __full_ieee754_pow(double,double);
#endif
#if defined(KRAIT_NEON_OPTIMIZATION)
int __ieee754_rem_pio2(double,double*) __attribute__((pcs("aapcs-vfp")));
double __kernel_sin(double,double,int) __attribute__((pcs("aapcs-vfp")));
double __kernel_cos(double,double) __attribute__((pcs("aapcs-vfp")));
double __kernel_tan(double,double,int) __attribute__((pcs("aapcs-vfp")));
int __kernel_rem_pio2(double*,double*,int,int,int,const int*) __attribute__((pcs("aapcs-vfp")));
#else
int __ieee754_rem_pio2(double,double*);
double __kernel_sin(double,double,int);
double __kernel_cos(double,double);
double __kernel_tan(double,double,int);
int __kernel_rem_pio2(double*,double*,int,int,int,const int*);
#endif
/* float versions of fdlibm kernel functions */
int __ieee754_rem_pio2f(float,float*);
float __kernel_sindf(double);
float __kernel_cosdf(double);
float __kernel_tandf(double,int);
int __kernel_rem_pio2f(float*,float*,int,int,int,const int*);
#endif /* !_MATH_PRIVATE_H_ */
|
