From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?utf-8?Q?Court=C3=A8s?=) Subject: Re: Porting to anything Date: Thu, 21 Feb 2013 10:36:40 +0100 Message-ID: <87d2vu9ffb.fsf@gnu.org> References: <8738yvnslh.fsf@karetnikov.org> <201302202024.22058.andreas@enge.fr> <87a9qyn461.fsf@gnu.org> <201302210038.58946.andreas@enge.fr> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([208.118.235.92]:57373) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1U8SaJ-0007ay-AY for bug-guix@gnu.org; Thu, 21 Feb 2013 04:37:05 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1U8SaC-0001bd-GF for bug-guix@gnu.org; Thu, 21 Feb 2013 04:36:58 -0500 Received: from [2a01:e0b:1:123:ca0a:a9ff:fe03:271e] (port=59200 helo=xanadu.aquilenet.fr) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1U8SaC-0001WV-9u for bug-guix@gnu.org; Thu, 21 Feb 2013 04:36:52 -0500 In-Reply-To: <201302210038.58946.andreas@enge.fr> (Andreas Enge's message of "Thu, 21 Feb 2013 00:38:58 +0100") 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: Andreas Enge Cc: bug-guix@gnu.org Andreas Enge skribis: > The currently unsatisfying situation, according to "HACKING", is that=20 > bootstrapping guix essentially starts by installing (and thus in the wors= t=20 > case porting) nix. Agreed. > I have spent the evening working on a shell script automating the=20 > compilation of the bootstrap binaries. Unsurprisingly, I end up=20 > reprogramming in shell part of make-bootstrap.scm. (In particular, it cos= t=20 > me some time to understand how to obtain statically linked binaries of th= e=20 > binutils.) And I am only halfway through, and maybe made some errors in t= he=20 > process that will only reveal themselves later. Isn=E2=80=99t it equally unsatisfying? :-) > So I wonder whether the dependency on nix is essential in make- > bootstrap.scm. Could one not renounce at working in a chroot with defined= =20 > inputs from nix, and execute the exact same code in the current=20 > environment, using the programs available in the current system from the= =20 > user's path? Then guix could be bootstrapped easily on any gnu/linux=20 > system. Yes. I outlined a plan in . However, I really think cross-compilation is the way to go. I have a tentative set of packages for the cross-toolchain that=E2=80=99s not fully = baked yet, and then we=E2=80=99ll have to adjust gnu-build-system to support cross-compilation. So that won=E2=80=99t happen overnight, but we could work toward that goal = for the next month or so. WDYT? Ludo=E2=80=99, who sympathizes with the feeling of frustration. :-/