From owner-freebsd-ports@FreeBSD.ORG Tue Jul 10 10:17:51 2012 Return-Path: Delivered-To: freebsd-ports@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 58237106566C; Tue, 10 Jul 2012 10:17:51 +0000 (UTC) (envelope-from dim@FreeBSD.org) Received: from tensor.andric.com (tensor.andric.com [87.251.56.140]) by mx1.freebsd.org (Postfix) with ESMTP id 106878FC12; Tue, 10 Jul 2012 10:17:51 +0000 (UTC) Received: from [IPv6:2001:7b8:3a7:0:f5a1:37a7:b01a:4353] (unknown [IPv6:2001:7b8:3a7:0:f5a1:37a7:b01a:4353]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) by tensor.andric.com (Postfix) with ESMTPSA id 2F6B65C37; Tue, 10 Jul 2012 12:17:44 +0200 (CEST) Message-ID: <4FFC0146.3030005@FreeBSD.org> Date: Tue, 10 Jul 2012 12:17:42 +0200 From: Dimitry Andric Organization: The FreeBSD Project User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:14.0) Gecko/20120619 Thunderbird/14.0 MIME-Version: 1.0 To: Doug Barton References: <4FF15A89.3000204@eskk.nu> <4FF2E349.5000202@FreeBSD.org> <20120704191351.GB70705@server.rulingia.com> <4FF49F51.8070600@FreeBSD.org> <4FFBD734.5030909@FreeBSD.org> <4FFBD7FB.3010002@FreeBSD.org> In-Reply-To: <4FFBD7FB.3010002@FreeBSD.org> X-Enigmail-Version: 1.5a1pre Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Peter Jeremy , freebsd-ports@FreeBSD.org, Leslie Jensen Subject: Re: make failed for editors/libreoffice X-BeenThere: freebsd-ports@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting software to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 10 Jul 2012 10:17:51 -0000 On 2012-07-10 09:21, Doug Barton wrote: > Thanks for looking into this. Did you see my message where I reported > that clang from ports works fine for LO on -current, but the base clang > fails? The lang/clang port is the same version as clang in base (3.1 release, minus a few patches needed for building world), but the port has asserts disabled by default, while clang in base is built with asserts enabled. Therefore, if you build LO with the clang port at its default settings, the assert that breaks the build is never triggered. However, I do not know if the resulting object code will have problems. > How does this change compare to the code that's already in the > clang port? This upstream change properly fixes the assert, which should fix the LO build with clang. It can also be applied to the lang/clang port, otherwise we'll have to wait until the 3.2 release to get it. Last but not least, to answer your question about the port not installing a clang-cpp link (or symlink): the clang installation scripts (either autoconf-based, or CMake-based) do not include such a link, which I think is an oversight by upstream. I'll lobby with them to include such a link in the future; meanwhile, maybe our port maintainer can manually add it. I added the clang-cpp link in the base system, because some parts of our build system depend on having an executable behaving as "cpp", instead of as "cc -E". The former preprocesses any file with unknown extension as if it were C, the latter doesn't.