https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114245
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |jakub at gcc dot gnu.org
--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
(In reply to Jonathan Wakely from comment #4)
> Updating the vtable is necessary during destruction if a later (i.e. less
> derived) destructor calls a virtual function. But if we can tell that the
> current dtor and all later base dtors are trivial, then we know that can't
> happen.
>
> If the front end detected that case, we wouldn't need to rely on dead store
> elimination.
But isn't that just an optimization the user asked not to perform (with using
-O0)?
I mean, e.g. the user could care about the vtable updates for e.g. calling
methods
from the debugger as the destruction of the object progresses.
