NAME

fenv — functions providing access to the floating-point environment

SYNOPSIS

#include <fenv.h>
#pragma STDC FENV_ACCESS ON

DESCRIPTION

The header <fenv.h> declares types, macros, and functions to provide access to the floating-point environment, consisting of any floating-point status flags and control modes supported by the implementation. A floating-point status flag is a variable whose value is set (but not cleared) when a floating-point exception is raised, which occurs as a side effect of floating-point arithmetic when an exceptional condition is encountered. A floating-point control mode is a variable whose value may be set by the user to affect the subsequent behavior of floating-point arithmetic.

Certain conventions are expected as part of the intended model for interaction with the floating-point environment:

• A function does not alter its caller's floating-point control modes, clear its caller's floating-point status flags, nor depend on the state of its caller's floating-point status flags unless the function is so documented.
• A function is assumed to require default floating-point control modes unless its documentation promises otherwise.
• A function is assumed to have the potential for raising floating-point exceptions unless its documentation promises otherwise.

Note that any usage of the functions provided by <fenv.h> should occur within the scope of

The FENV_ACCESS pragma provides a means to inform the compiler that the program might access the floating-point environment to test status flags or change the control modes. The pragma shall occur either outside external declarations (i.e. at file scope) or preceeding all explicit declarations and statements inside a compound statement. If part of a program tests floating-point status flags or sets control modes, but was translated with FENV_ACCESS "off", the behavior of that program is undefined.

The types fenv_t and fexcept_t represent the entire floating-point environment and the floating-point status flags, respectively. These should be treated as opaque types, and be manipulated only via the functions defined in <fenv.h>.

The macros FE_INVALID, FE_DIVBYZERO, FE_OVERFLOW, FE_UNDERFLOW, and FE_INEXACT expand to integer constant expressions such that bitwise ORs of any combination of the macros result in distinct values and bitwise ANDs of any combinations of the macros results in zero. These macros are used in functions defined in <fenv.h> to represent floating-point exceptions and status flags. Additionally, the macro FE_ALL_EXCEPT is defined to be the bitwise OR of all floating-point exception macros defined by the implementation.

The macros FE_TONEAREST, FE_TOWARDZERO, FE_DOWNWARD, and FE_UPWARD expand to integer constant expressions whose values are distinct nonnegative numbers. These are intended for use with the fesetround() and fegetround() functions for control of the floating-point rounding modes.

The macro FE_DFL_ENV represents the default floating-point environment--the one installed at program startup--and expands to a pointer to a const fenv_t object. It can be used as an argument to <fenv.h> functions that manage the floating-point environment.

The feclearexcept(), feraiseexcept(), fetestexcept(), fegetexceptflag(), and fesetexceptflag() functions provide access to the floating-point status flags.

The fegetround() and fesetround() functions provide control of rounding direction modes.

The fegetenv(), feholdexcept(), fesetenv(), and feupdateenv() functions provide control of the entire floating-point environment as an abstract object.

Further documentation for all of these functions is available in their respective manpages, and also in section 7.6 of the C standard.

SEE ALSO

feclearexcept(3), fegetenv(3), fegetround(3), fegetexceptflag(3), feholdexcept(3), fesetenv(3), fesetround(3), feraiseexcept(3), fesetexceptflag(3), feraiseexcept(3), feupdateenv(3)

STANDARDS

These functions conform to ISO/IEC 9899:TC3.