CKTIS / C++
Unary minus
Unary minus, written as -expression, computes the negated value. Learn how integral promotions affect the result, how unsigned negation works, and when signed overflow is a risk.
Version: C++23; repository compiler flag, production runtime not attested
Syntax
-expressionThe operator -x has one operand, unlike subtraction a - b. Built-in unary minus computes the negated value without changing x.
It accepts arithmetic types and unscoped enumerations. Integral and enumeration operands undergo integral promotions; the result has the promoted type. Consequently, -small in the example has type int. The result is a prvalue.
If the promoted type is unsigned, negation works modulo 2^N, where N is that type's width: -0u is 0u, and -1u is the largest unsigned int value. Do not assume the result will be negative.
Negating a signed integer type's minimum value causes undefined behavior when the result cannot fit in the promoted type, as with the minimum int value. For class types, an overloaded operator- can define different behavior; the rules above describe the built-in operator.
Examples
#include <iostream>
int main() {
int x = 7;
signed char small = 3;
auto negative = -small;
static_assert(sizeof(negative) == sizeof(int));
std::cout << -x << '\n';
std::cout << x << '\n';
std::cout << negative << '\n';
std::cout << -(-x) << '\n';
}
Result
-7 7 -3 7
The displayed code and declared result were verified in an isolated environment using the method recorded in the verification evidence for this exact version.