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
|
/*
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.
*/
/*
Simple, fast atan2 helper function for the fast trig routines by
Joachim Wesner <joachim.wesner@frankfurt.netsurf.de>, July 1998.
This routine should be compatible to gcc/egcs -O3 "inline" coding, where
calls to atan(x) are (for some strange ?? reason) replaced by calls to
__atan2(x, 1.0). That special case is detected and our atan(x) is
directly jumped to, instead of "dividing" by one. Otherwise our general
atan2() routine is used.
No special handling of illegal arguments or NANs, yet !!!
*/
.set noat
.set noreorder
#ifdef __ELF__
.section .rodata
#else
.rdata
#endif
.align 5
$C37:
.t_floating 1.0000000000000000e+00
.align 3
$C38:
.t_floating 3.1415926535897931e+00
.align 3
$C39:
.t_floating 1.5707963267948966e+00
.align 3
$C40:
.t_floating -1.5707963267948966e+00
.text
.align 5
.globl __atan2
.ent __atan2
__atan2:
ldgp $29,0($27)
lda $1, $C37
ldt $f10,0($1)
cmpteq $f10,$f17,$f1
fbeq $f1,$100
jmp $31,atan
.end __atan2
.align 5
.globl atan2
.ent atan2
atan2:
ldgp $29,0($27)
$100:
cmptlt $f31,$f17,$f1
fbeq $f1,$50
divt $f16,$f17,$f16
jmp $31,atan
.align 4
$50:
lda $30,-16($30)
.frame $30,16,$26,0
stq $26,0($30)
.mask 0x4000000,-16
.prologue 1
cmptlt $f17,$f31,$f1
fbeq $f1,$53
cmptle $f31,$f16,$f1
fbeq $f1,$54
divt $f16,$f17,$f16
jsr $26,atan
lda $1,$C38
ldt $f1,0($1)
addt $f0,$f1,$f0
br $31,$61
.align 4
$54:
divt $f16,$f17,$f16
jsr $26,atan
lda $1,$C38
ldt $f1,0($1)
subt $f0,$f1,$f0
br $31,$61
.align 4
$53:
cmptle $f31,$f16,$f1
lda $1,$C40
fbeq $f1,$59
lda $1,$C39
$59:
ldt $f0,0($1)
$61:
ldq $26,0($30)
addq $30,16,$30
ret $31,($26),1
.end atan2
|
