[Bug libstdc++/29286] [4.0/4.1/4.2/4.3 Regression] placement new does not change the dynamic type as it should

["gdr at cs dot tamu dot edu" <gcc-bugzilla at gcc dot gnu dot org>]

------- Comment #123 from gdr at cs dot tamu dot edu  2007-05-22 20:53 -------
Subject: Re:  [4.0/4.1/4.2/4.3 Regression] placement new does not change the
dynamic type as it should

"mrs at apple dot com" <gcc-bugzilla@gcc.gnu.org> writes:

| I think it is reasonable to push the tighening language into the
| standard, as that is what we meant and our wording was meant to be
| as tight as  C.

As C99, I guess.

I'm making the distinction between what the C++ standard currently
says, and what we would like it to say to benefit from some
optimizations with not too heroic effort.  That is why I would like a
core issue, and a resolution to change the current wording so that access
to union mambers are spelled out clearly.


| The effective type of an object for an access to its stored value is the
| declared type of the object, if any.

for C++, we need also to take into account placement-new (the original
subject reported here).  

-- Gaby


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29286