[Bug libstdc++/96074] New: Associative containers never propagate allocator on copy assignment

[Bug libstdc++/96074] New: Associative containers never propagate allocator on copy assignment

[alisdairm at me dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96074

            Bug ID: 96074
           Summary: Associative containers never propagate allocator on
                    copy assignment
           Product: gcc
           Version: 11.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libstdc++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: alisdairm at me dot com
  Target Milestone: ---

Created attachment 48834
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=48834&action=edit
Validate allocator propagation for all containers

The attached test driver is a simple validation that each container correctly
propgates allocators, given all 16 possible permutations of the traits.

All containers pass with the exception of the 4 associative containers that
each ignore the propagate_on_container_copy_assigment trait.

Note the test driver is slow to build as it instantiates a lot of templates.

[Bug libstdc++/96074] Associative containers never propagate allocator on copy assignment

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96074

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2020-07-06
             Status|UNCONFIRMED                 |NEW
     Ever confirmed|0                           |1

[Bug libstdc++/96074] Associative containers never propagate allocator on copy assignment

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96074

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |redi at gcc dot gnu.org
           Keywords|                            |wrong-code

--- Comment #1 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Off topic, in your test you have:

   // needed for gcc
   template <typename U>
   struct rebind { using other = allocator<U, COPY, MOVE, SWAP, false>; };

Isn't that needed always? allocator_traits can only provide a default rebind
for an allocator "of the form Alloc<U, Args>, where Args is zero or more type
arguments". Your non-type template arguments prevent that from working.

[Bug libstdc++/96074] Associative containers never propagate allocator on copy assignment

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96074

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |INVALID
             Status|ASSIGNED                    |RESOLVED

--- Comment #2 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Your test allocator has a bug. You do not override the is_always_equal trait
from the base class, which means you inherit std::allocator<T>::is_always_equal
which is defined to be std::true_type. So the associative containers elide the
propagation:

          if (_Alloc_traits::_S_propagate_on_copy_assign())
            {
              auto& __this_alloc = this->_M_get_Node_allocator();
              auto& __that_alloc = __x._M_get_Node_allocator();
              if (!_Alloc_traits::_S_always_equal()
                  && __this_alloc != __that_alloc)
                {
                  // Replacement allocator cannot free existing storage, we
need
                  // to erase nodes first.
                  clear();
                  std::__alloc_on_copy(__this_alloc, __that_alloc);
                }
            }

If I add this to the primary template and the partial specialization then all
tests pass (including the commented-out one for std::stringbuf):

   using is_always_equal                        = std::false_type;

You have been bitten by LWG 3170: https://cplusplus.github.io/LWG/issue3170

std::allocator<T>::is_always_equal is deprecated for precisely this reason
(it's a very annoying trap) but it's still there in C++20, so you need to
override it.

Closing, as the libstdc++ containers are already correct.