.\" Copyright (c) 2011, Apple Inc.
.Dd May 9, 2011
.Dt FECLEAREXCEPT 3
.Os OS X
.Sh NAME
.Nm feclearexcept, fegetexceptflag, feraiseexcept, fesetexceptflag, fetestexcept
.Nd functions providing access to the floating-point status flags.
.Sh SYNOPSIS
.Fd #include <fenv.h>
.Fd #pragma STDC FENV_ACCESS ON
.Ft int
.Fn feclearexcept "int excepts"
.Ft int
.Fn feraiseexcept "int excepts"
.Ft int
.Fn fetestexcept "int excepts"
.Ft int
.Fn fegetexceptflag "fexcept_t *flagp" "int excepts"
.Ft int
.Fn fesetexceptflag "fexcept_t *flagp" "int excepts"
.Sh DESCRIPTION
These functions provide access to the floating-point status flags.  The
.Vt int
input argument
.Fa excepts
for the functions represents a subset of floating-point exceptions, and can be
either zero or the bitwise OR of one or more floating-point exception macros
defined in <fenv.h>, for example
.Dv FE_OVERFLOW |
.Dv FE_INEXACT .
For other argument values the behavior of these functions is undefined.
.Pp
The
.Fn feclearexcept
function attempts to clear the supported floating-point flags corresponding
to the exceptions specified by its argument.  It returns zero if 
.Fa excepts
is zero or if the flags corresponding to all specified exceptions were
successfully cleared.  Otherwise, it returns a nonzero value.
.Pp
The
.Fn feraiseexcept
function attempts to raise the supported floating-point exceptions specified
by its argument.  Its effect is similar to that of arithmetic operations
raising the same exceptions; if traps are enabled for the exceptions that
are raised, they will be taken.  The order in which these exceptions are
raised is unspecified.  On OS X and iOS, raising overflow or underflow
using this function will additionally raise the inexact exception.
.Pp
The
.Fn feraiseexcept
function returns zero if
.Fa excepts
is zero or if the specified exceptions were successfully raised.  Otherwise a
nonzero value is returned.
.Pp
The
.Fn fetestexcept
function determines if any of the floating-point flags corresponding to the
exceptions specified by its argument are currently set.  It returns the bitwise
OR of the floating-point exception macros corresponding to the currently set
flags indicated by
.Fa excepts .
For example, if the underflow and inexact flags are set in the floating-point
environment, the result of
.Fn fetestexcept "FE_INEXACT | FE_INVALID"
will be
.Dv FE_INEXACT .
.Pp
The
.Fn fegetexceptflag
function attempts to store an implementation-defined representation of the
states of the floating-point status flags corresponding to the exceptions
specified by
.Fa excepts
in the object pointed to by the argument
.Fa flagp .
It returns zero if the representation is successfully stored, and a nonzero
value otherwise.
.Pp
The
.Fn fesetexceptflag
function attempts to set the floating-point status flags corresponding to the
exceptions specified by 
.Fa excepts
to the states stored in the object pointed to by
.Fa flagp .
This function does not raise floating-point exceptions--it only sets the state
of the flags.  The value of
.Fa *flagp
shall have been set by a previous call to
.Fn fegetexceptflag
whose second argument represented a superset of the exceptions represented by
.Fa excepts .
.Pp
The
.Fn fesetexceptflag
function returns zero if the
.Fa excepts
argument is zero or if all the specified flags were successfully set.  Otherwise
it returns a nonzero value.
.Sh SEE ALSO
.Xr fenv 3 ,
.Xr fegetenv 3 ,
.Xr fegetround 3 ,
.Xr feholdexcept 3 ,
.Xr fesetenv 3 ,
.Xr fesetround 3 ,
.Xr feupdateenv 3
.Sh STANDARDS
These functions conform to ISO/IEC 9899:TC3.