summaryrefslogtreecommitdiff
path: root/rts/StgPrimFloat.c
blob: f1f6736e0101f4f126e165d6083ef92d9f5859a8 (plain)
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
/* -----------------------------------------------------------------------------
 *
 * (c) Lennart Augustsson
 * (c) The GHC Team, 1998-2000
 *
 * Miscellaneous support for floating-point primitives
 *
 * ---------------------------------------------------------------------------*/

#include "PosixSource.h"
#include "Rts.h"

#include "StgPrimFloat.h"

#include <math.h>
#include <float.h>
#include <limits.h>

#define IEEE_FLOATING_POINT 1

#if FLT_RADIX != 2
# error FLT_RADIX != 2 not supported
#endif

/*
 * Encoding and decoding Doubles.  Code based on the HBC code
 * (lib/fltcode.c).
 */

#if IEEE_FLOATING_POINT
#define MY_DMINEXP  ((DBL_MIN_EXP) - (DBL_MANT_DIG) - 1)
/* DMINEXP is defined in values.h on Linux (for example) */
#define DHIGHBIT 0x00100000
#define DMSBIT   0x80000000

#define MY_FMINEXP  ((FLT_MIN_EXP) - (FLT_MANT_DIG) - 1)
#define FHIGHBIT 0x00800000
#define FMSBIT   0x80000000
#endif

#if defined(WORDS_BIGENDIAN) || defined(FLOAT_WORDS_BIGENDIAN)
#define L 1
#define H 0
#else
#define L 0
#define H 1
#endif

#define __abs(a)                (( (a) >= 0 ) ? (a) : (-(a)))

/** Trac #15271: Some large ratios are converted into double incorrectly.
  * This occurs when StgInt has 64 bits and C int has 32 bits, where wrapping
  * occurs and an incorrect signed value is passed into ldexp */
STATIC_INLINE int
truncExponent(I_ e)
{
#if INT_MAX < STG_INT_MAX
  if (RTS_UNLIKELY(e > INT_MAX))
    e = INT_MAX;
#endif
#if INT_MIN > STG_INT_MIN
  if (RTS_UNLIKELY(e < INT_MIN))
    e = INT_MIN;
#endif
  return e;
}

/* Special version for words */
StgDouble
__word_encodeDouble (W_ j, I_ e)
{
  StgDouble r;

  r = (StgDouble)j;

  /* Now raise to the exponent */
  if ( r != 0.0 ) /* Lennart suggests this avoids a bug in MIPS's ldexp */
    r = ldexp(r, truncExponent(e));

  return r;
}

/* Special version for small Integers */
StgDouble
__int_encodeDouble (I_ j, I_ e)
{
  StgDouble r;

  r = (StgDouble)__abs(j);

  /* Now raise to the exponent */
  if ( r != 0.0 ) /* Lennart suggests this avoids a bug in MIPS's ldexp */
    r = ldexp(r, truncExponent(e));

  /* sign is encoded in the size */
  if (j < 0)
    r = -r;

  return r;
}

/* Special version for small Integers */
StgFloat
__int_encodeFloat (I_ j, I_ e)
{
  StgFloat r;

  r = (StgFloat)__abs(j);

  /* Now raise to the exponent */
  if ( r != 0.0 ) /* Lennart suggests this avoids a bug in MIPS's ldexp */
    r = ldexp(r, truncExponent(e));

  /* sign is encoded in the size */
  if (j < 0)
    r = -r;

  return r;
}

/* Special version for small positive Integers */
StgFloat
__word_encodeFloat (W_ j, I_ e)
{
  StgFloat r;

  r = (StgFloat)j;

  /* Now raise to the exponent */
  if ( r != 0.0 ) /* Lennart suggests this avoids a bug in MIPS's ldexp */
    r = ldexp(r, truncExponent(e));

  return r;
}

/* This only supports IEEE floating point */

void
__decodeDouble_2Int (I_ *man_sign, W_ *man_high, W_ *man_low, I_ *exp, StgDouble dbl)
{
    /* Do some bit fiddling on IEEE */
    unsigned int low, high;             /* assuming 32 bit ints */
    int sign, iexp;
    union { double d; unsigned int i[2]; } u;   /* assuming 32 bit ints, 64 bit double */

    ASSERT(sizeof(unsigned int ) == 4            );
    ASSERT(sizeof(dbl          ) == 8            );
    ASSERT(sizeof(dbl          ) == SIZEOF_DOUBLE);

    u.d = dbl;      /* grab chunks of the double */
    low = u.i[L];
    high = u.i[H];

    if (low == 0 && (high & ~DMSBIT) == 0) {
        *man_low = 0;
        *man_high = 0;
        *exp = 0L;
    } else {
        iexp = ((high >> 20) & 0x7ff) + MY_DMINEXP;
        sign = high;

        high &= DHIGHBIT-1;
        if (iexp != MY_DMINEXP) /* don't add hidden bit to denorms */
            high |= DHIGHBIT;
        else {
            iexp++;
            /* A denorm, normalize the mantissa */
            while (! (high & DHIGHBIT)) {
                high <<= 1;
                if (low & DMSBIT)
                    high++;
                low <<= 1;
                iexp--;
            }
        }
        *exp = (I_) iexp;
        *man_low = low;
        *man_high = high;
        *man_sign = (sign < 0) ? -1 : 1;
    }
}

/* This is expected to replace uses of __decodeDouble_2Int() in the long run */
StgInt
__decodeDouble_Int64 (StgInt64 *const mantissa, const StgDouble dbl)
{
#if 0
    // We can't use this yet as-is, see ticket #9810
    if (dbl) {
        int exp = 0;
        *mantissa = (StgInt64)scalbn(frexp(dbl, &exp), DBL_MANT_DIG);
        return exp-DBL_MANT_DIG;
    } else {
        *mantissa = 0;
        return 0;
    }
#else
    I_ man_sign = 0;
    W_ man_high = 0, man_low = 0;
    I_ exp = 0;

    __decodeDouble_2Int (&man_sign, &man_high, &man_low, &exp, dbl);
    ASSIGN_Int64((W_*)mantissa, ((((StgInt64)man_high << 32)
                                  | (StgInt64)man_low)
                                 * (StgInt64)man_sign));
    return exp;
#endif
}

/* Convenient union types for checking the layout of IEEE 754 types -
   based on defs in GNU libc <ieee754.h>
*/

void
__decodeFloat_Int (I_ *man, I_ *exp, StgFloat flt)
{
    /* Do some bit fiddling on IEEE */
    int high, sign;                 /* assuming 32 bit ints */
    union { float f; int i; } u;    /* assuming 32 bit float and int */

    ASSERT(sizeof(int          ) == 4            );
    ASSERT(sizeof(flt          ) == 4            );
    ASSERT(sizeof(flt          ) == SIZEOF_FLOAT );

    u.f = flt;      /* grab the float */
    high = u.i;

    if ((high & ~FMSBIT) == 0) {
        *man = 0;
        *exp = 0;
    } else {
        *exp = ((high >> 23) & 0xff) + MY_FMINEXP;
        sign = high;

        high &= FHIGHBIT-1;
        if (*exp != MY_FMINEXP) /* don't add hidden bit to denorms */
            high |= FHIGHBIT;
        else {
            (*exp)++;
            /* A denorm, normalize the mantissa */
            while (! (high & FHIGHBIT)) {
                high <<= 1;
                (*exp)--;
            }
        }
        *man = high;
        if (sign < 0)
            *man = - *man;
    }
}