[Bug c++/126868] -ftrivial-auto-var-init=uninitialized should be incompatible with -std=c++26 (or at least warn about it)
"pinskia at gcc dot gnu.org via Gcc-bugs" <[email protected]>
| Newsgroups | gmane.comp.gcc.bugs |
|---|---|
| Message-ID | <[email protected]/bugzilla/> |
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126868
Drea Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |INVALID
Status|UNCONFIRMED |RESOLVED
--- Comment #7 from Drea Pinski <pinskia at gcc dot gnu.org> ---
>this is a *non-conforming* extension
-ffast-math and -fallow-store-data-races are also non-conforming extensions but
we don't warn when they are used. -fallow-store-data-races is nonconforming to
the C++11 memory model and can break some C++11+ code too. And can introduce
some race conditions that were not there before so it has just as much as a
security impact as -ftrivial-auto-var-init=uninitialized would have.
Since we already don't warn for -fallow-store-data-races there is no reason why
we should not warn about -ftrivial-auto-var-init=uninitialized.
Maybe some stronger wording in the manual would help but then again nobody
reads the documentation anyways.