From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?utf-8?Q?Court=C3=A8s?=) Subject: Re: [v2 1/1] doc: Show how to boot result of 'vm-image'. Date: Tue, 26 Jan 2016 22:11:07 +0100 Message-ID: <87zivsyrkk.fsf@gnu.org> References: <87powovek6.fsf@gnu.org> <20160126194215.GA19709@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]:56264) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aOAtO-00040C-C8 for guix-devel@gnu.org; Tue, 26 Jan 2016 16:11:15 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aOAtK-0004A9-Cw for guix-devel@gnu.org; Tue, 26 Jan 2016 16:11:14 -0500 In-Reply-To: <20160126194215.GA19709@jasmine> (Leo Famulari's message of "Tue, 26 Jan 2016 14:42:15 -0500") 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: Leo Famulari Cc: guix-devel@gnu.org Leo Famulari skribis: > On Tue, Jan 26, 2016 at 11:08:09AM +0100, Ludovic Court=C3=A8s wrote: [...] >> =E2=80=9CDefaults to 128@tie{}MiB, which may be insufficient for some >> operations.=E2=80=9D (It=E2=80=99s not the daemon specifically, it=E2= =80=99s mostly if you want >> to run X + Xfce, for instance. The former =E2=80=98guix substitute=E2= =80=99 used to >> take quite a lot of memory, but I think that=E2=80=99s no longer the cas= e since >> we switched to HTTP pipelining.) > > I hit this limit while using the "bare-bones" configuration template. > The system would boot with 128 MiB of RAM but any operations using the > daemon would fail. Oh OK. Well the thing is still using too much memory then (I suspect it=E2=80=99s =E2=80=98guix substitute=E2=80=99 that=E2=80=99s at fault, not= the C++ part of the daemon.) > Also, I looked into how QEMU interprets "megabyte" et al. From the QEMU > source file "include/qemu-common.h": > #define M_BYTE (1ULL << 20) > > So, I replaced "megabyte" with "mebibyte" in this paragraph. Perfect. :-) Thanks, Ludo=E2=80=99.