This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: objc failures on branch
Alexandre Oliva <aoliva@redhat.com> writes:
> On Mar 2, 2001, Andreas Jaeger <aj@suse.de> wrote:
>
> > Alexandre Oliva <aoliva@redhat.com> writes:
> >> On Feb 28, 2001, Andreas Jaeger <aj@suse.de> wrote:
> >>
> >> > I'm using a chroot that doesn't contain libgcc_s.so.0:
> >>
> >> > /usr/bin/ld: warning: libgcc_s.so.0, needed by /usr/src/packages/BUILD/gcc/obj/i486-suse-linux/libobjc/.libs/libobjc.so, not found (try using -rpath or -rpath-link)
> >>
> >> Then how come libobjc.so ended up depending on it?
>
> > Because it was build correctly:
>
> Err... You had said you didn't have libgcc_s.so.0 in the chrooted
> environment. Or I misunderstood.
What I wanted to say was: libgcc_s.so.0 was not installed in the
chroot environment - it exists only in the build directory but not in
the normal directories.
>
> Anyway, we do have a problem, indeed. libstdc++-v3 has worked around
> this by adding the appropriate -rpath flag when building test programs:
Exactly, that patch fixed it for libstdc++-v3.
> 2001-02-22 Benjamin Kosnik <bkoz@redhat.com>
>
> * tests_flags.in (CXXFLAGS): Add -rpath to gcc build dir.
>
> However, this assumes libtool is used to link the test programs
> (because -rpath isn't portable). What I don't understand is how come
> this works for check-g++, given that it doesn't use libtool for
> linking, AFAIK. Maybe it's just LD_LIBRARY_PATH? Or is there any
> hack in some of the g++-related .exp files that I can't seem to find
> right now?
I tried to figure this out myself - but couldn't find it either.:-(
But if you have a patch, I can easily test it.
Thanks for looking into this,
Andreas
--
Andreas Jaeger
SuSE Labs aj@suse.de
private aj@arthur.inka.de
http://www.suse.de/~aj