1fesetprec(3M)           Mathematical Library Functions           fesetprec(3M)
2
3
4

NAME

6       fesetprec, fegetprec - control floating point rounding precision modes
7

SYNOPSIS

9       c99 [ flag... ] file... -lm [ library... ]
10       #include <fenv.h>
11
12       int fesetprec(int prec);
13
14
15       int fegetprec(void);
16
17

DESCRIPTION

19       The IEEE 754 standard defines rounding precision modes for systems that
20       always deliver intermediate results to destinations in extended  double
21       precision  format.  These modes allow such systems to deliver correctly
22       rounded single and double precision results (in the absence  of  under‐
23       flow and overflow) with only one rounding.
24
25
26       The  fesetprec()  function  sets  the current rounding precision to the
27       precision specified by prec, which must be one of the following  values
28       defined in <fenv.h>:
29
30       FE_FLTPREC     round to single precision
31
32
33       FE_DBLPREC     round to double precision
34
35
36       FE_LDBLPREC    round to extended double precision
37
38
39
40       The default rounding precision when a program starts is FE_LDBLPREC.
41
42
43       The fegetprec() function returns the current rounding precision.
44

RETURN VALUES

46       The  fesetprec()  function  returns  a  non-zero value if the requested
47       rounding precision is established and 0 otherwise.
48

ATTRIBUTES

50       See attributes(5) for descriptions of the following attributes:
51
52
53
54
55       ┌───────────────────────────────────────────────────────────┐
56       │ATTRIBUTE TYPE                ATTRIBUTE VALUE              │
57       │Architecture                  Intel (see below)            │
58       │Availability                  SUNWlibms                    │
59       │Interface Stability           Stable                       │
60       │MT-Level                      MT-Safe                      │
61       └───────────────────────────────────────────────────────────┘
62
63
64       These functions are not available on SPARC systems because  SPARC  pro‐
65       cessors  deliver intermediate results to destinations in single or dou‐
66       ble format as determined by each floating point instruction.
67

SEE ALSO

69       fegetenv(3M), fesetround(3M), attributes(5)
70
71
72       Numerical Computation Guide
73
74
75
76SunOS 5.11                        12 Jul 2006                    fesetprec(3M)
Impressum