This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug c++/11495] New: [3.4 regression] Two-stage name lookup not Koenig-aware
- From: "bangerth at dealii dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 10 Jul 2003 21:08:01 -0000
- Subject: [Bug c++/11495] New: [3.4 regression] Two-stage name lookup not Koenig-aware
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11495
Summary: [3.4 regression] Two-stage name lookup not Koenig-aware
Product: gcc
Version: 3.4
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: c++
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: bangerth at dealii dot org
CC: gcc-bugs at gcc dot gnu dot org
Found while looking at 11493:
Another recent regression -- this code
------------------------
namespace NS {
struct C {};
void foo(C);
}
template <class T> void bar() { T t; foo (t); }
template void bar<NS::C> ();
-------------------------
should be legal: foo(t) is a dependent call, so foo should be looked up
at instantiation time, and then Koenig lookup applies to find NS::foo.
In fact, recent gccs compiled this just fine, but with tonight's snapshot,
it doesn't anymore:
tmp/gg> ../build-gcc/gcc-install/bin/c++ -c x.cc
x.cc: In function `void bar() [with T = NS::C]':
x.cc:8: instantiated from here
x.cc:6: error: `foo' undeclared (first use this function)
x.cc:6: error: (Each undeclared identifier is reported only once for each
function it appears in.)
:256: error: `<declaration error>' is not a function,
x.cc:3: error: conflict with `void NS::foo(NS::C)'
x.cc:6: error: in call to `foo'
What is also pretty funny is the third to last line. This usually indicates
a built-in function, but how can a builtin have a declaration error? And
how can it conflict?
W.