From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?utf-8?Q?Court=C3=A8s?=) Subject: Re: Cross compiling libxml2 Date: Thu, 12 Dec 2013 22:46:02 +0100 Message-ID: <8738lxagkl.fsf@gnu.org> References: <20131212180550.GA4062@intra> 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]:46072) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VrE5D-0000hX-23 for guix-devel@gnu.org; Thu, 12 Dec 2013 16:46:17 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VrE56-0007Wt-Si for guix-devel@gnu.org; Thu, 12 Dec 2013 16:46:10 -0500 Received: from mail2-relais-roc.national.inria.fr ([192.134.164.83]:37457) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VrE56-0007Wn-Lr for guix-devel@gnu.org; Thu, 12 Dec 2013 16:46:04 -0500 In-Reply-To: <20131212180550.GA4062@intra> (John Darrington's message of "Thu, 12 Dec 2013 19:05:52 +0100") List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org To: John Darrington Cc: guix-devel@gnu.org John Darrington skribis: > But another problem which I'm not sure how to fix is that (assoc-ref inpu= ts "libc") > returns false. Hence the build fails: Actually, when cross-compiling, the phase procedures are passed #:inputs and #:native-inputs. So =E2=80=9Clibc=E2=80=9D is in the #:native-inputs a= list, while =E2=80=9Ccross-libc=E2=80=9D is in the #:inputs alist. So you=E2=80=99d have to do something like that: (assoc-ref inputs ,(if (%current-target-system) "cross-libc" "libc")) This is not great, and we should probably fix that in the next =E2=80=98core-updates=E2=80=99. In the meantime, that=E2=80=99s what we ha= ve. The packages needed to cross-build =E2=80=98bootstrap-tarballs=E2=80=99 (as= Hydra does) either do not refer explicitly to libc, or have a special case that removes the need to refer to it (as for ncurses), which is why we never came across that before. Thanks, Ludo=E2=80=99.