errno

From cppreference.com
< cpp‎ | error
 
 
 
Defined in header <cerrno>
#define errno /* implementation-defined */

errno is a preprocessor macro used for error indication. It expands to a static(until C++11)thread-local(since C++11) modifiable lvalue of type int.

Several standard library functions indicate errors by writing positive integers to errno. Typically, the value of errno is set to one of the error codes, listed in <cerrno> as macro constants that begin with the letter E, followed by uppercase letters or digits.

The value of errno is 0 at program startup, and although library functions are allowed to write positive integers to errno whether or not an error occurred, library functions never store 0 in errno.

Example

#include <cerrno>
#include <clocale>
#include <cmath>
#include <cstring>
#include <iostream>
 
int main()
{
    const double not_a_number = std::log(-1.0);
    std::cout << not_a_number << '\n';
 
    if (errno == EDOM)
    {
        std::cout << "log(-1) failed: " << std::strerror(errno) << '\n';
        std::setlocale(LC_MESSAGES, "de_DE.utf8");
        std::cout << "Or, in German, " << std::strerror(errno) << '\n';
    }
}

Possible output:

nan
log(-1) failed: Numerical argument out of domain
Or, in German, Das numerische Argument ist ausserhalb des Definitionsbereiches

Defect reports

The following behavior-changing defect reports were applied retroactively to previously published C++ standards.

DR Applied to Behavior as published Correct behavior
LWG 310 C++98 it was unclear whether errno is a macro
or an identifier with external linkage
errno must
be a macro

See also

macros for standard POSIX-compatible error conditions
(macro constant)
displays a character string corresponding of the current error to stderr
(function)
returns a text version of a given error code
(function)
C documentation for errno