https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80328
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |jakub at gcc dot gnu.org
--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Yeah, -fexcess-precision= is not an option for C++, but still, if you use
-ffloat-store, you need to be prepared for severe performance degradation. If
you have several levels of abstraction as you have in the std::array case, you
need to accept that those abstractions might introduce some extra stores.
