This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: GCC 3.0 Status Report
- To: mark at codesourcery dot com (Mark Mitchell)
- Subject: Re: GCC 3.0 Status Report
- From: Joe Buck <jbuck at racerx dot synopsys dot com>
- Date: Thu, 19 Apr 2001 17:42:08 -0700 (PDT)
- Cc: dave at hiauly1 dot hia dot nrc dot ca, chastain at cygnus dot com, law at redhat dot com, gcc at gcc dot gnu dot org, libstdc++ at gcc dot gnu dot org, gcc-patches at gc dot gnu dot org
Mark writes:
> + // Build don't link:
> + // Origin: Mark Mitchell <mark@codesourcery.com>
> +
> + namespace N {
> + extern "C" int i;
> +
> + void f () {
> + i = 3;
> + }
> + };
> +
> + int i;
> +
> + int main () { N::f (); }
Perhaps I'm missing something, but if you say "Build don't link"
it seems that the test does not catch the error. If We get an .o
file that incorrectly references N::i, it's still a valid .o file.
The program must be linked to reveal the undefined symbol. Right?