std::fegetround, std::fesetround

C++

Language

Headers

Freestanding and hosted implementations

Named requirements

Language support library

Concepts library (C++20)

Diagnostics library

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)

Filesystem library (C++17)

Technical Specifications

Numerics library

Common mathematical functions

Special mathematical functions

Floating-point environment (C++11)

Complex numbers

Numeric arrays

Pseudo-random number generation

Compile-time rational arithmetic (C++11)

Numeric algorithms

gcd (C++17)

lcm (C++17)

Generic numeric operations

iota (C++11)

accumulate

inner_product

adjacent_difference

partial_sum

reduce (C++17)

transform_reduce (C++17)

inclusive_scan (C++17)

exclusive_scan (C++17)

transform_inclusive_scan (C++17)

transform_exclusive_scan (C++17)

Bit operations

bit_cast (C++20)

ispow2 (C++20)

ceil2 (C++20)

floor2 (C++20)

log2p1 (C++20)

Floating-point environment

Functions

feclearexcept (C++11)

fetestexcept (C++11)

feraiseexcept (C++11)

fegetexceptflagfesetexceptflag (C++11)(C++11)

fegetroundfesetround (C++11)(C++11)

fegetenvfesetenv (C++11)

feholdexcept (C++11)

feupdateenv (C++11)

Macro constants

FE_ALL_EXCEPTFE_DIVBYZEROFE_INEXACTFE_INVALIDFE_OVERFLOWFE_UNDERFLOW (C++11)

FE_DOWNWARDFE_TONEARESTFE_TOWARDZEROFE_UPWARD (C++11)

FE_DFL_ENV (C++11)

Defined in header <cfenv>
int fesetround( int round )	(1)	(since C++11)
int fegetround()	(2)	(since C++11)

Manages the floating-point rounding direction.

1) Attempts to establish the floating-point rounding direction equal to the argument round, which is expected to be one of the floating point rounding macros.

2) Returns the value of the floating point rounding macro that corresponds to the current rounding direction.

Parameters

round

-

rounding direction, one of floating point rounding macros

Return value

1) 0 on success, non-zero otherwise.

2) the floating point rounding macro describing the current rounding direction or a negative value if the direction cannot be determined

Notes

The current rounding mode, reflecting the effects of the most recent fesetround, can also be queried with FLT_ROUNDS

See floating-point rounding macros for the effects of rounding.

Example

Run this code

#include <cmath>
#include <cfenv>
#include <iostream>

int main()
{
    #pragma STDC FENV_ACCESS ON
    std::fesetround(FE_DOWNWARD);
    std::cout << "rounding using FE_DOWNWARD: \n" << std::fixed
              << " 12.0 ->  " << std::nearbyint(12.0) << '\n'
              << " 12.1 ->  " << std::nearbyint(12.1) << '\n'
              << "-12.1 -> " << std::nearbyint(-12.1) << '\n'
              << " 12.5 ->  " << std::nearbyint(12.5) << '\n'
              << " 12.9 ->  " << std::nearbyint(12.9) << '\n'
              << "-12.9 -> " << std::nearbyint(-12.9) << '\n'
              << " 13.0 ->  " << std::nearbyint(13.0) << '\n';
    std::fesetround(FE_TONEAREST);
    std::cout << "rounding using FE_TONEAREST: \n"
              << " 12.0 ->  " << std::nearbyint(12.0) << '\n'
              << " 12.1 ->  " << std::nearbyint(12.1) << '\n'
              << "-12.1 -> " << std::nearbyint(-12.1) << '\n'
              << " 12.5 ->  " << std::nearbyint(12.5) << '\n'
              << " 12.9 ->  " << std::nearbyint(12.9) << '\n'
              << "-12.9 -> " << std::nearbyint(-12.9) << '\n'
              << " 13.0 ->  " << std::nearbyint(13.0) << '\n';
}

Output:

rounding using FE_DOWNWARD:
 12.0 ->  12.000000
 12.1 ->  12.000000
-12.1 -> -13.000000
 12.5 ->  12.000000
 12.9 ->  12.000000
-12.9 -> -13.000000
 13.0 ->  13.000000
rounding using FE_TONEAREST: 
 12.0 ->  12.000000
 12.1 ->  12.000000
-12.1 -> -12.000000
 12.5 ->  12.000000
 12.9 ->  13.000000
-12.9 -> -13.000000
 13.0 ->  13.000000

See also

nearbyint (C++11)	nearest integer using current rounding mode (function)
rintlrintllrint (C++11)(C++11)(C++11)	nearest integer using current rounding mode with exception if the result differs (function)