From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?utf-8?Q?Court=C3=A8s?=) Subject: Re: Failed to set up the build environment =?utf-8?Q?for=E2=80=A6?= Date: Sat, 22 Mar 2014 15:17:47 +0100 Message-ID: <87zjkixqpg.fsf@gnu.org> References: <87txaue9dp.fsf@serenity.home> <87k3bq6vj1.fsf@gnu.org> <20140320143802.GA22707@debian> <87fvmcycac.fsf@gnu.org> <20140320211554.GA14256@debian> <87ob10v9a3.fsf@gnu.org> <20140321123120.GA20520@debian.eduroam.u-bordeaux.fr> <87zjkj4sy0.fsf@gnu.org> <20140321134449.GA5037@debian> 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]:55180) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WRMkF-00009y-Fu for guix-devel@gnu.org; Sat, 22 Mar 2014 10:18:01 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WRMk9-0001Cw-LL for guix-devel@gnu.org; Sat, 22 Mar 2014 10:17:55 -0400 Received: from hera.aquilenet.fr ([2a01:474::1]:44489) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WRMk9-0001CK-Dw for guix-devel@gnu.org; Sat, 22 Mar 2014 10:17:49 -0400 In-Reply-To: <20140321134449.GA5037@debian> (Andreas Enge's message of "Fri, 21 Mar 2014 14:44:49 +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: Andreas Enge Cc: guix-devel Andreas Enge skribis: > On Fri, Mar 21, 2014 at 01:49:59PM +0100, Ludovic Court=C3=A8s wrote: >> For now, the =E2=80=9Csolution=E2=80=9D is to create it like this: >> sudo mknod /dev/kvm c 10 232 >> Can you confirm? > > Yes, that works. Together with /dev/shm (how is this affected by the chan= ge, > by the way?), there are more and more things to change on a debian system. > Hopefully your patch will go through. Could /dev/kvm not simply be droppe= d, > instead of enabled according to the machine on which the daemon runs? Then > we would be closer to a deterministic build also. We can=E2=80=99t simply drop it, because we want to be able to use KVM in t= he build environment. That=E2=80=99s a minor issue, though, because (1) the only things that rely= on KVM are the VM build support, and (2) when the stand-alone GNU exists, it will have /dev/kvm. Ludo=E2=80=99.