[Bug c++/71069] -Waddress didn't catch all cases

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71069

Martin Sebor <msebor at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |diagnostic
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2016-07-06
                 CC|                            |msebor at gcc dot gnu.org
     Ever confirmed|0                           |1
           Severity|normal                      |enhancement

--- Comment #4 from Martin Sebor <msebor at gcc dot gnu.org> ---
I agree that diagnosing these problems at compile time would be ideal, but it's
not always possible (in fact, I suspect it rarely is: only when GCC can prove
that the reference is initialized by dereferencing a null pointer).

What I think could help, though, is warning for the test that GCC eliminates
based on the assumption that the reference address will never be null. 
Unfortunately, to do that in any but the most trivial cases, the warning would
need to rely on optimization.  Still, I'll confirm this as an enhancement
request to extend -Waddress to help highlight at least some of these problems.