Re: nasty link problem with Gfortran and G++ (C++) ???

On Tue, 07 Jan 2025 15:35:38 -0600, Lynn McGuire wrote:

I just ran into a nasty problem with GFortran and G++ (C++).  Probably
not a bug ???  I am using GCC 14.1.
 
I have a lot of code in C++ (over 100,000 lines).  I have 750,000 lines
of code in GFortran.  I have to extern "C" this code in C++ to make it
callable by GFortran code.
 
I missed declaring a couple of C++ functions as extern "C" which means
that they kept their C++ mangled link names.  But these C++ functions
were declared as integer*8 and external in the GFortran code (old F77
code).
 
So, the GCC linker did not report to me that it did not have a link for
the G++ functions.  This may be a bug, I am not sure.
 
When I ran the program, the C++ functions were not called by the
GFortran code.  Instead, the GFortran code treated the C++ functions as
integer*8 scalar variables.
 
If needful, I can probably put together a small code sample that
exhibits the problem.  Maybe.  It could be that the size of my code
affects the GCC linker.
 
I removed the Gfortran code "external" keywords, extern "C" the C++
functions, added the C++ function to my module list, and got a working
link.

Need to see some actual code.

% cat > foo.c
int
foo(void)
{
   return (1);
}
% gcc14 -c foo.c
% nm foo.o
0000000000000000 T foo
% g++14 -c foo.c
% nm foo.o
0000000000000000 T _Z3foov
% cat > foo.f90
integer*8 function foo()
   foo = 1
end
% gfortran14 -c foo.f90
% nm foo.o
0000000000000000 T foo_

So, with gcc, you get the symbol foo. With g++, you get the mangled
named _Z3foov.  With gfortran, you get foo_.



--
steve