FE_DFL_ENV

 

C++

Language

Concepts

Utilities library

Strings library

Containers library

Algorithms library

Iterators library

Numerics library

Input/output library

Localizations library

Regular expressions library (C++11)

Atomic operations library (C++11)

Thread support library (C++11)

 

Numerics library

Common mathematical functions

Floating-point environment

Complex numbers

Numeric arrays

Pseudo-random number generation

Compile-time rational arithmetic (C++11)

Generic numeric operations

iota (C++11)

accumulate

inner_product

adjacent_difference

partial_sum

 

Floating-point environment

feclearexcept (C++11)

fetestexcept (C++11)

feraiseexcept (C++11)

fegetexceptflag fesetexceptflag (C++11) (C++11)

fegetround (C++11)

fesetround (C++11)

fegetenv (C++11)

feholdexcept (C++11)

fesetenv (C++11)

feupdateenv (C++11)

FE_ALL_EXCEPT FE_DIVBYZERO FE_INEXACT FE_INVALID FE_OVERFLOW FE_UNDERFLOW (C++11)

FE_DOWNWARD FE_TONEAREST FE_TOWARDZERO FE_UPWARD (C++11)

FE_DFL_ENV (C++11)

 

Defined in header <cfenv>
#define FE_DFL_ENV  /*implementation defined*/		(since C++11)

The macro constant FE_DFL_ENV expands to an expression of type const std::fenv_t*, which points to a full copy of the default floating-point environment, that is, the environment as loaded at program startup.

Additional macros that begin with FE_ followed by uppercase letters, and have the type const std::fenv_t*, may be supported by an implementation.

Example

#include <iostream>
#include <cfenv>
 
#pragma STDC FENV_ACCESS ON
 
void show_env()
{
    int e = std::fetestexcept(FE_ALL_EXCEPT);
    if(e & FE_DIVBYZERO) std::cout << "division by zero is raised\n";
    if(e & FE_INEXACT)   std::cout << "inexact is raised\n";
    if(e & FE_INVALID)   std::cout << "invalid is raised\n";
    if(e & FE_UNDERFLOW) std::cout << "underflow is raised\n";
    if(e & FE_OVERFLOW)  std::cout << "overflow is raised\n";
    int r = std::fegetround();
    switch(r)
    {  
        case FE_DOWNWARD: std::cout << "rounding down\n"; break;
        case FE_TONEAREST: std::cout << "rounding to nearest\n"; break;
        case FE_TOWARDZERO: std::cout << "rounding to zero\n"; break;
        case FE_UPWARD: std::cout << "rounding up\n"; break;
    }
}
 
int main()
{
    std::cout << "On startup: \n";
    show_env();
 
    std::feraiseexcept(FE_UNDERFLOW | FE_OVERFLOW);
    std::fesetround(FE_UPWARD);
 
    std::cout << "\nBefore restoration: \n";
    show_env();
 
    std::fesetenv(FE_DFL_ENV);
 
    std::cout << "\nAfter reset to default: \n";
    show_env();
}

Output:

On startup: 
rounding to nearest
 
Before restoration: 
underflow is raised
overflow is raised
rounding up
 
After reset to default: 
rounding to nearest

See also

fegetenv fesetenv (C++11)	saves or restores the current floating point environment (function)
feupdateenv (C++11)	restores the floating-point environment and raises the previously raise exceptions (function)