This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: Fix PR 35942: remove -lstdc++ from libtool postdeps for CXX.
- From: "John Z. Bohach" <jzb2 at aexorsyst dot com>
- To: Ralf Wildenhues <Ralf dot Wildenhues at gmx dot de>
- Cc: gcc-patches at gcc dot gnu dot org, libstdc++ at gcc dot gnu dot org, earthengine at gmail dot com, bonzini at gnu dot org
- Date: Sat, 16 Jan 2010 08:25:54 -0800
- Subject: Re: Fix PR 35942: remove -lstdc++ from libtool postdeps for CXX.
- References: <20100107200233.GD10034@gmx.de> <20100116132619.GD5460@gmx.de>
Thank you for fixing this...I confirmed this back on
Comment #12 From John Z. Bohach 2009-02-09 20:25
on the PR.
--john
On Saturday 16 January 2010 05:26:19 am Ralf Wildenhues wrote:
> Ping <http://gcc.gnu.org/ml/gcc-patches/2010-01/msg00349.html>
>
> Note that I haven't gotten any feedback on whether this actually
> fixes PR 35942.
>
> Thanks,
> Ralf
>
> * Ralf Wildenhues wrote on Thu, Jan 07, 2010 at 09:02:33PM CET:
> > Bootstrapped and regtested on x86_64-unknown-linux-gnu, i.e., only
> > native build. There has been no feedback yet whether this patch
> > fixes the PR, but I'm fairly confident. OK for trunk when/if that
> > is confirmed?
> >
> > Fix PR 35942: remove -lstdc++ from libtool postdeps for CXX.
> >
> > libstdc++-v3/ChangeLog:
> > 2010-01-07 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
> >
> > PR libstdc++/35942
> > * configure.ac: Remove -lstdc++ from libtool's postdeps_CXX.
> > * configure: Regenerate.
> >
> > diff --git a/libstdc++-v3/configure.ac b/libstdc++-v3/configure.ac
> > index 2ee2bdb..22c830b 100644
> > --- a/libstdc++-v3/configure.ac
> > +++ b/libstdc++-v3/configure.ac
> > @@ -98,6 +98,9 @@ AM_PROG_LIBTOOL
> > AC_SUBST(enable_shared)
> > AC_SUBST(enable_static)
> >
> > +# Eliminate -lstdc++ addition to postdeps for cross compiles.
> > +postdeps_CXX=`echo " $postdeps_CXX " | sed 's, -lstdc++ ,,g'`
> > +
> > # Possibly disable most of the library.
> > ## TODO: Consider skipping unncessary tests altogether in this
> > case, rather ## than just ignoring the results. Faster /and/ more
> > correct, win win.