[Bug c++/53012] unrelated friend operators in same namespace interfere with operator resolution outside of namespace

[alphaetapi at hotmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53012

Adam H. Peterson  changed:

   What|Removed |Added

 CC||alphaetapi at hotmail dot
   ||com

--- Comment #1 from Adam H. Peterson  2012-04-16 
23:22:47 UTC ---
I get the same results here.  I've also determined that the problem doesn't
rely on using operator overloading or #include-ing .  Here's a
program that fails similarly without using either (and in fact has no
#includes):


namespace one {
class A { };
}
void func(const ::one::A& ) { }

namespace two {
class unrelated_a {
friend void func( const unrelated_a& );
};
#if not defined MAKE_IT_WORK
class unrelated_b {
friend void func( const unrelated_b& );
};
#endif

void output( const one::A& val ) {
func(val);
}
}
int main() {
two::output( one::A() );
}



With the above code, I get "'func' was not declared in this scope" inside
output()'s body.  While it's true that func() is not declared in namespace two,
it is in the global scope.  And in fact, apparently it is found if the
definition for class unrelated_b is removed.

[Bug c++/69078] [C++14] function local static not initialized when only used in a generic/variadic lambda

[alphaetapi at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69078

Adam H. Peterson  changed:

   What|Removed |Added

 CC||alphaetapi at hotmail dot com

--- Comment #1 from Adam H. Peterson  ---
I'm seeing this too, in one of my work projects.  And I also see the odd
results when the variable is used both in the lambda and outside of it.

#include 
template
void run( F &&f ) {
f(nullptr);
}
struct V {
int i;
};
int main() {
static V const s={2};
std::cout << s.i << "\n";
run([](auto){
std::cout << s.i << "\n" ;
});
}

Expected output:
2
2
Actual output:
2
0

(The in-function use seems to have been inlined, while the lambda continues to
access an uninitialized state.)

[Bug tree-optimization/26854] Inordinate compile times on large routines

[alphaetapi at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=26854

Adam H. Peterson  changed:

   What|Removed |Added

 CC||alphaetapi at hotmail dot com

--- Comment #137 from Adam H. Peterson  ---
Incidentally, my GCC version is 5.3:

$ g++ --version
g++ (Debian 5.3.1-13) 5.3.1 20160323
Copyright (C) 2015 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

My OS is Debian/Testing:
$ uname -a
Linux -desktop 4.4.0-1-amd64 #1 SMP Debian 4.4.6-1 (2016-03-17)
x86_64 GNU/Linux

[Bug c++/69078] [C++14] function local static not initialized when only used in a generic/variadic lambda

[alphaetapi at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69078

--- Comment #2 from Adam H. Peterson  ---
Incidentally, my GCC version is 5.3:

$ g++ --version
g++ (Debian 5.3.1-13) 5.3.1 20160323
Copyright (C) 2015 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

My OS is Debian/Testing:
$ uname -a
Linux -desktop 4.4.0-1-amd64 #1 SMP Debian 4.4.6-1 (2016-03-17)
x86_64 GNU/Linux

[Bug tree-optimization/26854] Inordinate compile times on large routines

[alphaetapi at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=26854

--- Comment #138 from Adam H. Peterson  ---
Disregard my comment above.  I was dropped into the wrong bug.