From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.windriver.com (mail.windriver.com [147.11.1.11]) by mail.openembedded.org (Postfix) with ESMTP id 129A070252 for ; Fri, 11 Jul 2014 01:18:45 +0000 (UTC) Received: from ALA-HCA.corp.ad.wrs.com (ala-hca.corp.ad.wrs.com [147.11.189.40]) by mail.windriver.com (8.14.5/8.14.5) with ESMTP id s6B1IlTm017984 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for ; Thu, 10 Jul 2014 18:18:47 -0700 (PDT) Received: from [128.224.162.181] (128.224.162.181) by ALA-HCA.corp.ad.wrs.com (147.11.189.40) with Microsoft SMTP Server id 14.3.169.1; Thu, 10 Jul 2014 18:18:46 -0700 Message-ID: <53BF3B75.6020607@windriver.com> Date: Fri, 11 Jul 2014 09:18:45 +0800 From: Robert Yang User-Agent: Mozilla/5.0 (X11; Linux i686; rv:24.0) Gecko/20100101 Thunderbird/24.6.0 MIME-Version: 1.0 To: References: <1404886727-44130-1-git-send-email-liezhi.yang@windriver.com> <20140709095544.GC2446@jama> <53BD17A1.6000205@windriver.com> In-Reply-To: Subject: Re: [meta-oe] [PATCH] fftwf and fftwl: exclude them from world X-BeenThere: openembedded-devel@lists.openembedded.org X-Mailman-Version: 2.1.12 Precedence: list Reply-To: openembedded-devel@lists.openembedded.org List-Id: Using the OpenEmbedded metadata to build Distributions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Jul 2014 01:18:53 -0000 Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit On 07/10/2014 05:49 PM, Koen Kooi wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Robert Yang schreef op 09-07-14 12:21: >> >> >> On 07/09/2014 05:55 PM, Martin Jansa wrote: >>> On Tue, Jul 08, 2014 at 11:18:47PM -0700, Robert Yang wrote: >>>> The fftw, fftwf and fftwl provide the same files (different >>>> contents), we should only build one of them, otherwise there would be >>>> warning >>>> >>>> WARNING: The recipe fftw is trying to install files into a shared >>>> area when those files already exist. Those files and their manifest >>>> location are: sysroots/qemux86-64/usr/include/fftw3.f03 Matched in >>>> manifest-qemux86-64-fftwl.populate_sysroot >>>> manifest-qemux86-64-fftwf.populate_sysroot >>>> sysroots/qemux86-64/usr/include/fftw3.h Matched in >>>> manifest-qemux86-64-fftwl.populate_sysroot >>>> manifest-qemux86-64-fftwf.populate_sysroot >>>> sysroots/qemux86-64/usr/include/fftw3q.f03 Matched in >>>> manifest-qemux86-64-fftwl.populate_sysroot >>>> manifest-qemux86-64-fftwf.populate_sysroot >>>> sysroots/qemux86-64/usr/include/fftw3.f Matched in >>>> manifest-qemux86-64-fftwl.populate_sysroot >>>> manifest-qemux86-64-fftwf.populate_sysroot >>>> sysroots/qemux86-64/usr/include/fftw3l.f03 Matched in >>>> manifest-qemux86-64-fftwl.populate_sysroot >>>> manifest-qemux86-64-fftwf.populate_sysroot Please verify which >>>> package should provide the above files. >>>> >>>> and maybe errors (race issue): >>>> >>>> The recipe imagemagick.do_compiler error: fourier.c:69:19: fatal >>>> error: fftw3.h: No such file or directory #include >>>> >>>> This is because imagemagick.do_configure depends on >>>> fftw.populate_sysroot, but the files installed by >>>> fftw.populate_sysroot, maybe removed by fftwl or fftwf when they >>>> rebuild. >>>> >>>> Only build fftw in the world will fix the problem. >>> >>> But it doesn't fix the problem for people who are really building >>> fftwl and fftw. If there is someone with this use-case can he send >>> patch moving common parts to separate recipe used by all 3 or just >>> rename/move files to have unique name? >>> >>> Or if they aren't needed at the same time, lets convert fftwl and >>> fftwf to PACKAGECONFIG options in fftw. >> >> Yes, I think that PACKAGECONFIG is a good idea, will send patches >> tomorrow if no objections. > > The original idea of the different packages was that you could have all the > libraries in the binary feeds since they are needed. With packageconfig that > becomes virtually impossible. > I don't have a suggestion to improve the current situation, though :( The race issue is painful since we don't know when it would happen, I think that exclude two of them from world can reduce the race issue for most of the users, for the user who build or install 3 of them for the same image, we don't know how to fix it, but we already have the warning for that. // Robert > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.5 (Darwin) > Comment: GPGTools - http://gpgtools.org > > iD8DBQFTvmG+MkyGM64RGpERAu1zAJsGXdlezcHjV4JFjhTj9zeSaJ271ACdERQE > Xye89UlN8wZ3HEmV0fNGYMQ= > =iZI9 > -----END PGP SIGNATURE----- >