From mboxrd@z Thu Jan 1 00:00:00 1970 From: Federico Beffa Subject: Re: [PATCH 4/4] gnu: matplotlib: Add gtk3 backends. Date: Tue, 23 Dec 2014 18:09:30 +0100 Message-ID: References: <878ui8edzl.fsf@gnu.org> <87vblahv6n.fsf@gnu.org> <878uhzkynx.fsf@gnu.org> 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]:49898) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Y3Sxg-0003pz-J5 for guix-devel@gnu.org; Tue, 23 Dec 2014 12:09:33 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Y3Sxf-0005sO-NF for guix-devel@gnu.org; Tue, 23 Dec 2014 12:09:32 -0500 In-Reply-To: <878uhzkynx.fsf@gnu.org> 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: =?UTF-8?Q?Ludovic_Court=C3=A8s?= Cc: Guix-devel On Tue, Dec 23, 2014 at 1:17 AM, Ludovic Court=C3=A8s wrote: > I found that =E2=80=9Cguix build python2-matplotlib -n=E2=80=9D triggers = a seemingly > infinite loop (either uses of =E2=80=98package-with-python2=E2=80=99 some= how introduce > cycles, or they lead to very large DAGs), which is what is causing Hydra > evaluation failures. > > I tried reverting 25f9a0 but it doesn=E2=80=99t help. > > Could you try to bisect it and revert or fix the problem? If it turns > out to require more time, could you just comment out the offending parts > of python.scm so Hydra can resume? The build does finish as I've tested it on my machine before committing the package. However, for some reason, guix needs very long to generate the derivation. On my machine (quad-core Xeon E5520) python2-matplotlib takes ca. 31 mins to start building. Python2-scipy, which includes python2-matplotlib, takes ca. 80 mins.! Currently I do not understand the reason, but I note that matplotlib, numpy and scipy are the only python packages with a large number of inputs. I've tried to use trace in the REPL and I see that there is a very large activity about inputs and their origins. However, the buffer just becomes too large to handle. It needs more sophisticated debugging methods. For the moment I've commented out python2-matplotlib and the python-2 versions of the packages making use of it: scipy and numpy. Sorry for the inconvenience! I did not expect that this would cause troubles to hydra. Fede