Re: using C11's _Atomic()

On Mon, May 4, 2026 at 3:40 AM Tony Cook <[email protected]> wrote:

> On Mon, May 04, 2026 at 02:25:34AM +0200, Leon Timmermans wrote:
> > That can be worked around by using std::atomic<int> instead of
> _Atomic(int)
> > in C++ mode (and including atomic).
>
> I couldn't find anything that indicated they were guaranteed to be
> compatible - does an _Atomic<long long> have the same ABI as
> atomic<long long>?
>
> They do provide all the same type aliases though, so if they aren't
> compatible C++ code that uses <atomic> and <stdatomic.h> is going to
> be painful.
>
> I saw one claim that it is UB:
>
>
> https://stackoverflow.com/questions/53898429/interoperabilty-between-c-and-c-atomics#answer-53898931
>
> but they might be wrong.


Strictly it's probably UB, but practically any lock-free atomic type should
compile down to its underlying type as its representation in either case,
so I'm not overly worried until I've seen proof otherwise.

Leon