From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marius Bakke Subject: `guix copy` vs grafts Date: Thu, 28 Dec 2017 01:40:49 +0100 Message-ID: <87efnfd6r2.fsf@fastmail.com> References: <87h8sdwd5z.fsf@elephly.net> <87r2rfddrw.fsf@fastmail.com> Mime-Version: 1.0 Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha512; protocol="application/pgp-signature" Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:48136) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eUMFn-0002za-V5 for help-guix@gnu.org; Wed, 27 Dec 2017 19:41:02 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eUMFk-0006zN-6o for help-guix@gnu.org; Wed, 27 Dec 2017 19:40:59 -0500 Received: from out2-smtp.messagingengine.com ([66.111.4.26]:49617) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1eUMFi-0006y8-Tm for help-guix@gnu.org; Wed, 27 Dec 2017 19:40:55 -0500 In-Reply-To: <87r2rfddrw.fsf@fastmail.com> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: help-guix-bounces+gcggh-help-guix=m.gmane.org@gnu.org Sender: "Help-Guix" To: Ricardo Wurmus , help-guix --=-=-= Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Marius Bakke writes: > Ricardo Wurmus writes: > >> Hi Guix, >> >> as my i686 laptop has only about 1G of memory and thus fails to build a >> great many packages (including the guix package itself) I wanted to >> build its configuration on my x86_64 laptop. I then copied the system >> closure to the i686 laptop with =E2=80=9Cguix copy=E2=80=9D. >> >> On the i686 target I ran =E2=80=9Csudo -E guix system reconfigure config= .scm=E2=80=9D >> and expected it to do little more than switching to the transferred >> system item. Instead it started building Guix, which fails on my i686 >> machine (not enough memory). > > There is a caveat with `guix copy`: if you copy the grafted derivation > (as the `guix system build` closure would be) the receiving system won't > recognize it since it only computes the ungrafted derivations. To clarify: I think what happens is that "guix system/build" builds the 'normal' derivation, then afterwards computes and builds the grafts. If the first is not present locally, it will be built even if the grafted derivation is already in the store. --=-=-= Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQEzBAEBCgAdFiEEu7At3yzq9qgNHeZDoqBt8qM6VPoFAlpEPZEACgkQoqBt8qM6 VPpjQwgAirflp6rSNlaPcXJ0Qe3jWKITCIOpqnL8uYWQZq293gTTvLI8jhwkITzj BWVhB507MDGmhWUiO+w7oK++wVrESgAZP6a7++NOswr8pj7pXxZb99Ahkr4e5tlp 2lpGYINmNpZQx0kNSpOfAFRqrlxh0qHCg9xBW776bzLDwbJ08dryeBifTwq/6lvD F6NB5nz9njWx5SNCwmFMAYkCLh3GZ5RQulnyyTbmlGJx/3tPE0raueZ+ShRK4Bk3 CJIE+VA4AcA7J4YoErxZF2m7tUaHs+PjAP4c/px9C3HxhhdAUS7TlStzbNoG2qGE xGN7PDSlhA/jdhAvgMZ6JP8buL8i1g== =Pync -----END PGP SIGNATURE----- --=-=-=--