This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug c++/13358] long long and C++ do not mix well in 3.3/3.4
- From: "lloyd at randombit dot net" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 8 Dec 2003 19:01:22 -0000
- Subject: [Bug c++/13358] long long and C++ do not mix well in 3.3/3.4
- References: <20031208180553.13358.lloyd@randombit.net>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Additional Comments From lloyd at randombit dot net 2003-12-08 19:01 -------
| Then their compiler is broken and you have to report the problem
| there.
Yes. Wonderful. GCC is being gratuitously incompatible in this case,
especially considering that such a constant was SILENTLY accepted by
3.[012]*, even in C++, even with -Wall. And now it produces an error
which can only be turned into a warning with -fpermissive, which
also lets in all kinds of things I *would* want to know about.
And I have yet to see where in C99 that "0xFEDCBA9876543210" with
no suffix is not a valid long long constant. If I am truly stupid
and this is not legal in C99, please put me out of my misery, give
me the section number where it says this, and I will slink away in
shame.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13358