From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?UTF-8?Q?Court=C3=A8s?=) Subject: bug#22659: Collision of /bin/ld Date: Sun, 21 Feb 2016 23:17:32 +0100 Message-ID: <87fuwlso77.fsf@gnu.org> References: <20160214091227.GA10462@jasmine> <87twl9pbas.fsf@netris.org> <20160216195923.GB19703@jasmine> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:35564) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aXcKL-0001U8-2L for bug-guix@gnu.org; Sun, 21 Feb 2016 17:18:05 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aXcKI-0002my-Ur for bug-guix@gnu.org; Sun, 21 Feb 2016 17:18:04 -0500 Received: from debbugs.gnu.org ([208.118.235.43]:39514) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aXcKI-0002mt-Qm for bug-guix@gnu.org; Sun, 21 Feb 2016 17:18:02 -0500 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84) (envelope-from ) id 1aXcKI-0007SU-Ea for bug-guix@gnu.org; Sun, 21 Feb 2016 17:18:02 -0500 Sender: "Debbugs-submit" Resent-Message-ID: In-Reply-To: (Chris Marusich's message of "Wed, 17 Feb 2016 17:35:48 +0000") List-Id: Bug reports for GNU Guix List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: bug-guix-bounces+gcggb-bug-guix=m.gmane.org@gnu.org Sender: bug-guix-bounces+gcggb-bug-guix=m.gmane.org@gnu.org To: Chris Marusich Cc: 22659@debbugs.gnu.org, 22659-done@debbugs.gnu.org Chris Marusich skribis: >> This collision is expected. The 'ld' within ld-wrapper, generated from >> the template in gnu/packages/ld-wrapper.in, automatically adds -rpath >> arguments to the linker for each shared library, so that the runtime >> linker will be able to find them in their non-standard locations. > > If one of those 'ld' files is the real ld, and the other is supposed to be > a wrapper around the real one that adds functionality, then how is it that > things will always work out OK if only one of them is installed? I'm not > sure how the "arbitrary" selection works, but doesn't the choice matter? The choice matters, indeed, but it=E2=80=99s deterministic: the first one w= ins. And in this case, the first one is the wrapper. Now, it might be best for =E2=80=98guix environment=E2=80=99 to silence the= se warnings. Ludo=E2=80=99.