From owner-freebsd-questions@FreeBSD.ORG Tue Mar 25 09:19:17 2014 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id AFA529AD for ; Tue, 25 Mar 2014 09:19:17 +0000 (UTC) Received: from blue.qeng-ho.org (blue.qeng-ho.org [217.155.128.241]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mx1.freebsd.org (Postfix) with ESMTPS id 2EFB3F2A for ; Tue, 25 Mar 2014 09:19:16 +0000 (UTC) Received: from fileserver.home.qeng-ho.org (localhost [127.0.0.1]) by fileserver.home.qeng-ho.org (8.14.7/8.14.5) with ESMTP id s2P9JE7R006532; Tue, 25 Mar 2014 09:19:15 GMT (envelope-from freebsd@qeng-ho.org) Message-ID: <53314A12.4010502@qeng-ho.org> Date: Tue, 25 Mar 2014 09:19:14 +0000 From: Arthur Chance User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:24.0) Gecko/20100101 Thunderbird/24.4.0 MIME-Version: 1.0 To: Assaf Gordon , freebsd-questions@freebsd.org Subject: Re: 'make' behavior in FreeBSD-10 ? References: <532F1A58.30702@wi.mit.edu> <44siq7wjwi.fsf@be-well.ilk.org> <53307B1E.4080704@wi.mit.edu> In-Reply-To: <53307B1E.4080704@wi.mit.edu> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 25 Mar 2014 09:19:17 -0000 On 24/03/2014 18:36, Assaf Gordon wrote: > Hello, > > On 03/24/2014 02:06 PM, Lowell Gilbert wrote: >> >>> I'm trying to track down an issue in compiling 'autotools' packages >>> that appears in FreeBSD-10 . >>> (I'm not a frequent FreeBSD user, so it's possible I'm missing >>> something - any help would be appreciated). >> >> Is automake specifically what you're trying to build, or are the other >> questions relevant as well? >> >> This is relevant because it sounds as though you are not trying to use >> the FreeBSD ports (or pkg; they're really the same thing in terms of >> what gets installed) system. That's the usual way such a thing would be >> done, and the exceptions are almost entirely people who would be able to >> answer such questions for themselves. > > I'm trying to build a package that uses autotools+automake, and should > be built with "./confgure && make". > > These issues were detected in an package of mine, and were reproducible > with the canonical "GNU-Hello" project > ( https://www.gnu.org/software/hello/ ) . > > I'd assume that almost any project that uses "automake" would faces > these issues. > > I am not using "pkg" - under the (wrong?) assumption that a portable > package written in C and uses "autotools" should be portable enough to > be compiled on many different OSes. It would be better to say that a portable package written in C and using autotools should be portable enough to be compiled on many different distros of Linux and possibly on other systems if they have sufficient of the Gnu tools installed. autotools is a Gnu software suite and presumes a Gnu infrastructure including Gnu make. The initial BSD release which was the ancestor of all the *BSDs today was six years before Gnu was even announced as a concept. >> >>> It is my understanding the FreeBSD-10 switched to a different 'make' >>> (bmake?), which could be the reason for these differences - but I do >>> not know how to fix them. >> >> Gnu Make has never been standard on BSD. Most variations of make >> programs understand slightly different makefile syntax, so they >> generally don't run each other's makefiles. > > I understand that there are inherent differences. > > To rephrase my previous question, to a very narrow technical issue: > Is there a way to make such a "makefile" complete successfully on FreeBSD: > === > all: > non_existing_command ; true > === > > Or will this construct never work on FreeBSD, and must be changed in > order to be portable ? > Because as a whole, it is a valid shell command, which should succeed > (return exit code 0), according to POSIX (if I understand correctly). > > > Thank you, > -gordon