From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?utf-8?Q?Court=C3=A8s?=) Subject: Re: Compilation failure on very large module Date: Tue, 13 Jun 2017 14:18:52 +0200 Message-ID: <87fuf42i8j.fsf@gnu.org> References: 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]:60851) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dKkmg-0001CU-Qe for guix-devel@gnu.org; Tue, 13 Jun 2017 08:18:59 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dKkmc-0007ZW-Sp for guix-devel@gnu.org; Tue, 13 Jun 2017 08:18:58 -0400 In-Reply-To: (Ricardo Wurmus's message of "Tue, 13 Jun 2017 11:11:53 +0200") 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" To: Ricardo Wurmus Cc: guix-devel Hello, Ricardo Wurmus skribis: > Here=E2=80=99s the crash with Guile 2.2. I=E2=80=99ve added the module t= o the list of > modules that should be compiled and ran =E2=80=9Cmake=E2=80=9D. You can = see that there > are a couple of undefined variables, but I don=E2=80=99t think this shoul= d lead > to a crash. Looks like the same problem as outlined here: . Basically the compiler=E2=80=99s memory consumption seems to be proportiona= l to the size of the input file, when I think it should be proportional to the complexity of the most complex top-level expression given that Guile doesn=E2=80=99t do inter-procedural analysis. (The failure you get on 2.0 seems to be another story: apparently we=E2=80= =99re hitting a limitation somewhere in the object format.) Ludo=E2=80=99.