blob: c7858c2dceb9ec9568b04bdcb8d99eb569399ba6 (
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
|
@node machine,,syscalls,Top
@chapter NEC V70 Specific Functions
The NEC V70 has machine instructions for fast IEEE floating-point
arithmetic, including operations normally provided by the library.
When you use the @file{/usr/include/fastmath.h} header file, the
names of several library math functions are redefined to call the
@code{fastmath} routine (using the corresponding V70 machine instructions)
whenever possible.
For example,
@example
#include <fastmath.h>
double sqsin(x)
double x;
@{
return sin(x*x);
@}
@end example
expands into the code
@example
@dots{}
double sqsin(x)
double x;
@{
return fast_sin(x*x);
@}
@end example
The library has an entry @code{fast_sin} which uses the machine
instruction @code{fsin.l} to perform the operation. Note that the
built-in instructions cannot set @code{errno}
in the same way that the C coded functions do. Refer to a V70
instruction manual to see how errors are generated and handled.
Also, the library provides true @code{float} entry points. The
@code{fast_sinf} entry point really performs a @code{fsin.s}
operation. Because of this, the instructions are only useful when
using code compiled with an ANSI C compiler. The prototypes
and definitions for the floating-point versions of the math library
routines are only defined if compiling a module with an ANSI C
compiler.
@page
@section Entry points
The functions provided by @file{fastmath.h} are
@example
double fast_sin(double); /* fsin.l */
double fast_cos(double); /* fcos.l */
double fast_tan(double); /* ftan.l */
double fast_asin(double); /* fasin.l */
double fast_acos(double); /* facos.l */
double fast_atan(double); /* fatan.l */
double fast_sinh(double); /* fsinh.l */
double fast_cosh(double); /* fcosh.l */
double fast_tanh(double); /* ftanh.l */
double fast_asinh(double); /* fasinh.l */
double fast_acosh(double); /* facosh.l */
double fast_atanh(double); /* fatanh.l */
double fast_fabs(double); /* fabs.l */
double fast_sqrt(double); /* fsqrt.l */
double fast_exp2(double); /* fexp2.l */
double fast_exp10(double); /* fexp10.l */
double fast_expe(double); /* fexpe.l */
double fast_log10(double); /* flog10.l */
double fast_log2(double); /* flog2.l */
double fast_loge(double); /* floge.l */
float fast_sinf(float); /* fsin.s */
float fast_cosf(float); /* fcos.s */
float fast_tanf(float); /* ftan.s */
float fast_asinf(float); /* fasin.s */
float fast_acosf(float); /* facos.s */
float fast_atanf(float); /* fatan.s */
float fast_sinhf(float); /* fsinh.s */
float fast_coshf(float); /* fcosh.s */
float fast_tanhf(float); /* ftanh.s */
float fast_asinhf(float); /* fasinh.s */
float fast_acoshf(float); /* facosh.s */
float fast_atanhf(float); /* fatanh.s */
float fast_fabsf(float); /* fabs.s */
float fast_sqrtf(float); /* fsqrt.s */
float fast_exp2f(float); /* fexp2.s */
float fast_exp10f(float); /* fexp10.s */
float fast_expef(float); /* fexpe.s */
float fast_log10f(float); /* flog10.s */
float fast_log2f(float); /* flog2.s */
float fast_logef(float); /* floge.s */
@end example
|
