https://gcc.gnu.org/bugzilla/show_bug.cgi?id=124143
--- Comment #11 from Eric Botcazou <ebotcazou at gcc dot gnu.org> --- > I'm looking at this again and questioning why SP_Node exists at all. I don't > see a reason to not put its members directly on the subpool type which would > save an allocation and would mean that GNAT follows the implementation > advice in A.18.32(15/5) (specifically "Bounded holder objects should be > implemented without dynamic allocation"). Yes, and allocating something on the heap seems questionable in the context.
