[Bug libstdc++/67707] [5/6 regression] Regression in std::move of std::deque with non-is_always_equal allocator

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67707

--- Comment #2 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Ugh, I hate allocators.

I probably should have added a completely new std::__cxx11::deque for gcc-5
with nothrow move and full allocator support, instead of trying to bolt it on
to our existing std::deque.

Testing this fix:

--- a/libstdc++-v3/include/bits/stl_deque.h
+++ b/libstdc++-v3/include/bits/stl_deque.h
@@ -644,6 +644,7 @@ _GLIBCXX_BEGIN_NAMESPACE_CONTAINER
        _Tp_alloc_type __sink __attribute((__unused__)) {std::move(__alloc)};
        // Create an empty map that allocates using the moved-from allocator.
        _Deque_base __empty{__alloc};
+       __empty._M_initialize_map(0);
        // Now safe to modify current allocator and perform non-throwing swaps.
        _Deque_impl __ret{std::move(_M_get_Tp_allocator())};
        _M_impl._M_swap_data(__ret);