https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115955
--- Comment #3 from Jonathan Wakely <redi at gcc dot gnu.org> --- (In reply to Jonathan Wakely from comment #1) > Yes, this is a known limitation that will get fixed with the planned > refactoring of atomic::wait. The function that maps an address to a mutex has been moved out of the headers for gcc 16, so can be changed to use a better hash function without causing an ABI break. That makes GCC 16 incompatible with GCC 15, of course, but that was always going to be true because we need ABI changes in order to get C++20 stable and non-experimental.
