std::atomic_ref<T>::operator+=,-=,&=,|=,^=

member only of atomic_ref<Integral> and atomic_ref<Floating> template specializations 
T operator+=( T arg ) const noexcept;
 (1)
member only of atomic_ref<T*> template specialization 
T* operator+=( std::ptrdiff_t arg ) const noexcept;
 (1)
member only of atomic_ref<Integral> and atomic_ref<Floating> template specializations 
T operator-=( T arg ) const noexcept;
 (1)
member only of atomic_ref<T*> template specialization 
T* operator-=( std::ptrdiff_t arg ) const noexcept;
 (1)
member only of atomic_ref<Integral> template specialization 
T operator&=( T arg ) const noexcept;
 (3) 
T operator|=( T arg ) const noexcept;
 (4) 
T operator^=( T arg ) const noexcept;
 (5)

Atomically replaces the current value of the referenced object with the result of computation involving the previous value and arg. These operations are read-modify-write operations.

1) Performs atomic addition. Equivalent to return fetch_add(arg) + arg;.
2) Performs atomic subtraction. Equivalent to return fetch_sub(arg) - arg;.
3) Performs atomic bitwise and. Equivalent to return fetch_and(arg) & arg;.
4) Performs atomic bitwise or. Equivalent to return fetch_or(arg) | arg;.
5) Performs atomic bitwise exclusive or. Equivalent to return fetch_xor(arg) ^ arg;.

For signed integral types, arithmetic is defined to use two’s complement representation. There are no undefined results.

For floating-point types, the floating-point environment in effect may be different from the calling thread's floating-point environment. The operation need not be conform to the corresponding std::numeric_limits traits but is encouraged to do so. If the result is not a representable value for its type, the result is unspecified but the operation otherwise has no undefined behavior.

For T* types, the result may be an undefined address, but the operations otherwise have no undefined behavior. The program is ill-formed if T is not an object type.

Parameters

arg - the argument for the arithmetic operation

Return value

The resulting value (that is, the result of applying the corresponding binary operator to the value immediately preceding the effects of the corresponding member function).

Notes

Unlike most compound assignment operators, the compound assignment operators for atomic_ref do not return a reference to their left-hand arguments. They return a copy of the stored value instead.

See also

atomically increments or decrements the referenced object by one
(public member function)

© cppreference.com
Licensed under the Creative Commons Attribution-ShareAlike Unported License v3.0.
http://en.cppreference.com/w/cpp/atomic/atomic_ref/operator_arith2