https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90348
Alexander Monakov <amonakov at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |amonakov at gcc dot gnu.org
--- Comment #17 from Alexander Monakov <amonakov at gcc dot gnu.org> ---
I think part of the problem is trying to make "deaths" explicit via CLOBBERs
without making "births" also explicit in the IR. Doing both would have allowed
a lifetime verifier that checks that births dominate all references to a
variable and likewise deaths/clobbers postdominate all references, which would
presumably catch this early and make the IR more rigorous overall.
