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
|
/*
libffm - Free, pretty fast replacement for some math (libm) routines
on Linux/AXP, optimized for the 21164
Copyright (C) 1998 Joachim Wesner <joachim.wesner@frankfurt.netsurf.de>
and Kazushige Goto <goto@statabo.rim.or.jp>
This 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.
This 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 this library (see file COPYING.LIB); if not, write
to the Free Software Foundation, Inc., 675 Mass Ave, Cambridge,
MA 02139, USA.
*/
/*
Fast log approximation(s) including range reduction by Joachim Wesner,
joachim.wesner@frankfurt.netsurf.de, see also mc 8/1991 p. 78-93.
July 12 1998 JW.
No special handling of illegal arguments or NANs, yet !!!
*/
/* improved and re-scheduled by Kazushige Goto <goto@statabo.rim.or.jp> */
.set noat
.set noreorder
#ifdef __ELF__
.section .rodata
#else
.rdata
#endif
.align 5
R:
.t_floating 7.07106781186547572737e-1 # SQRT05
.t_floating 1.00000000000000000000e0
.t_floating 6.93147180559945286227e-1 # LOG2
.t_floating 3.01029995663981198017e-1 # LOG10_2
/* Derived from Chebyshev Approx. */
/* rel. error 1.5e-18 */
.t_floating 2.62332293764593771357e-1
.t_floating 2.13182032931477888349e-1
.t_floating 4.12198585366190917156e-1
.t_floating 9.61796693925982992823e-1
.t_floating 3.20598577851792521098e-1
.t_floating 2.20962847584267652046e-1
.t_floating 5.77078016348200772967e-1
.t_floating 2.88539008177792677401e0
.text
.align 5
.globl log2
.ent log2
log2:
lda $30, -16($30)
ldgp $29,.-log2($27)
stt $f16, 0($30)
.frame $30,16,$26,0
#ifdef PROF
lda $28, _mcount
jsr $28, ($28), _mcount
unop
unop
#endif
.prologue 1
lda $2, R
ldt $f22, 0($2) # R[0]
ldt $f1, 8($2)
lda $3, 8($2) # LOG_X
br $31, $continue
.end log2
.align 5
.globl log10
.ent log10
log10:
lda $30, -16($30)
ldgp $29,.-log10($27)
stt $f16, 0($30)
.frame $30,16,$26,0
#ifdef PROF
lda $28, _mcount
jsr $28, ($28), _mcount
unop
unop
#endif
.prologue 1
lda $2, R
ldt $f22, 0($2) # R[0]
ldt $f1, 8($2)
lda $3, 24($2) # LOG_X
br $31, $continue
.end log10
.align 5
.globl log
.ent log
log:
lda $30, -16($30)
ldgp $29,.-log($27)
stt $f16, 0($30)
.frame $30,16,$26,0
#ifdef PROF
lda $28, _mcount
jsr $28, ($28), _mcount
unop
unop
#endif
.prologue 1
lda $2, R
ldt $f22, 0($2) # R[0]
ldt $f1, 8($2)
lda $3, 16($2) # LOG_X
.align 4
$continue:
cpyse $f22, $f16, $f11 # copy E
cmptlt $f11, $f22, $f10
subt $f11, $f1, $f29 # y4 = x - R[1]
addt $f11, $f1, $f30 # y6 = x + R[1]
fbeq $f10, $34 # if (x<y)
addt $f29, $f11, $f29 # y4 += x
addt $f30, $f11, $f30 # y6 += x
$34:
divt $f29, $f30, $f18
# wait, wait, wait, for a long time
ldq $1, 0($30)
ldt $f23, 32($2) # R[4]
ldt $f27, 64($2) # R[8]
srl $1, 52, $1
lda $1,-1022($1)
fbeq $f10, $35 # if (x<y)
subl $1,1,$1
$35:
stq $1,8($30)
ldt $f24, 40($2) # R[5]
ldt $f25, 48($2) # R[6]
ldt $f26, 56($2) # R[7]
ldt $f28, 72($2) # R[9]
ldt $f29, 80($2) # R[10]
ldt $f30, 88($2) # R[11]
ldt $f15, 8($30) # load iexp
addq $30, 16, $30
mult $f18, $f18, $f19 # y = x * x
mult $f19, $f19, $f20 # y2 = y * y
mult $f23, $f19, $f23 # t2 = R[4] * y
mult $f24, $f19, $f24 # t1 = R[5] * y
mult $f25, $f19, $f25 # t3 = R[6] * y
mult $f20, $f20, $f21 # y4 = y2 * y2
mult $f26, $f19, $f26 # t4 = R[7] * y
addt $f23, $f27, $f23 # t2 += R[8]
addt $f24, $f28, $f24 # t1 += R[9]
addt $f25, $f29, $f25 # t3 += R[10]
addt $f26, $f30, $f26 # t4 += R[11]
mult $f21, $f20, $f22 # y6 = y4 * y2
ldt $f1, 0($3)
mult $f23, $f21, $f23 # t2 *= y4
mult $f24, $f22, $f24 # t1 *= y6
mult $f25, $f20, $f25 # t3 *= y2
cvtqt $f15, $f0 # int -> float
addt $f24, $f23, $f23 # t1 += t2
addt $f23, $f25, $f23 # t1 += t3
addt $f23, $f26, $f23 # t1 += t4
mult $f23, $f18, $f23 # y *= x
addt $f23, $f0, $f0 # y + iexp
mult $f0, $f1, $f0
ret $31,($26),1
.end log
|
