Received: from smarthost.techfak.uni-bielefeld.de ([129.70.137.17]) by linuxtogo.org with esmtp (Exim 4.72) (envelope-from ) id 1QvQA5-0005Q5-Cv for openembedded-devel@lists.openembedded.org; Mon, 22 Aug 2011 10:47:13 +0200 Received: from [129.70.144.11] (hooge.TechFak.Uni-Bielefeld.DE [129.70.144.11]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: sherbrec) by smarthost.TechFak.Uni-Bielefeld.DE (Postfix) with ESMTPSA id 76450AC for ; Mon, 22 Aug 2011 10:42:29 +0200 (CEST) Message-ID: <4E521642.9040304@cit-ec.uni-bielefeld.de> Date: Mon, 22 Aug 2011 10:41:38 +0200 From: Stefan Herbrechtsmeier User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0) Gecko/20110812 Thunderbird/6.0 MIME-Version: 1.0 To: openembedded-devel@lists.openembedded.org References: <20110818131043.GC26690@golden> <1313689684.7918.2.camel@eng-phousel-ubuntu> <4E4E2DFB.5050106@cit-ec.uni-bielefeld.de> <4E4E62AB.1050603@balister.org> <4E4E7416.1070105@gmail.com> <4E4E964A.6090607@cit-ec.uni-bielefeld.de> <20110819191222.GA3730@golden> In-Reply-To: <20110819191222.GA3730@golden> Subject: Re: [beagleboard] Unable to build Angstrom meta-toolchain X-BeenThere: openembedded-devel@lists.openembedded.org X-Mailman-Version: 2.1.11 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: Mon, 22 Aug 2011 08:47:13 -0000 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: quoted-printable Am 19.08.2011 21:12, schrieb fabian b=FCttner: > On 18:58 Fri 19 Aug , Stefan Herbrechtsmeier wrote: >> Am 19.08.2011 16:32, schrieb Khem Raj: >>> On 8/19/2011 6:18 AM, Philip Balister wrote: >>>> On 08/19/2011 02:33 AM, Stefan Herbrechtsmeier wrote: >>>>> Am 18.08.2011 19:48, schrieb Peter Housel: >>>>>> On Thu, 2011-08-18 at 15:10 +0200, fabian b=FCttner wrote: >>>>>>> Hi, >>>>>>> >>>>>>> I am having troubles compiling the Angstrom SDK. >>>>>>> >>>>>>> I used the angstrom-setup-scripts to configure OE, sourced the >>>>>>> environment settings and built the base-image/console-image with >>>>>>> bitbake. >>>>>>> >>>>>>> While trying to compile the meta-toolchain, just when bitmake >>>>>>> finishes >>>>>>> downloading task-sdk-bare_1.0-r2.6_armv7a.ipk, I am getting an er= ror >>>>>>> output like this: >>>>>>> >>>>>>> http://pastie.org/2390409 >>>>>>> >>>>>>> After that, I tried building the meta-toolchain(i did a bitmake c= lean >>>>>>> meta-toolchain) without '-j2' but the error I receive is ultimate= ly >>>>>>> the same: >>>>>>> * satisfy_dependencies_for: Cannot satisfy the following dependen= cies >>>>>>> * for task-sdk-bare: >>>>>>> * libstdc++-dev * >>>>>>> * opkg_install_cmd: Cannot install package task-sdk-bare. >>>>>>> >>>>>>> Unfortunately I am too unexperienced with OE/Angstrom to fix bugs >>>>>>> like that. >>>>>>> Maybe someone could give me a hint how to get around this >>>>>>> show-stopper? >>>>>> I can confirm that I've been seeing this as well on recent >>>>>> 2011.03-maintenance, but haven't had a chance to troubleshoot the = root >>>>>> cause yet. >>>>>> >>>>> It was inbtroduced with commit 105d881227d06f31c180dffb777ee63361b2= f2b8 >>>>> "gcc: Package libstdc++ gdb python helpers into dev package". >>>>> If no helpers exist the empty package will not be generated. >>>>> I have try to add >>>>> ALLOW_EMPTY_libstdc++-dev =3D "1" >>>>> to gcc-package-cross.inc but this breaks meta-toolchain as the file= s in >>>>> ${prefix}/${TARGET_SYS}/usr/include/c++ are missing. >>>>> >>>>> Only removing libstdc++-dev package from gcc-package-cross.inc fix = the >>>>> problem. >>>> Does anyone know if dv has the same problem? I can't check easily at= the >>>> moment. >>> what happens if you remove libstdc++-dev libgcc-dev from PACKAGES in >>> recipes/gcc/gcc-package-cross.inc ? >> Maybe I was not precise enough above, but removing libstdc++-dev from >> PACKAGES >> in recipes/gcc/gcc-package-cross.inc and bump INC_PR in gcc-4.3.3.inc >> fix the problems. > Hi Stefan, > thanks for your fix, as soon as I finish building another base-image, I > am going to try your suggestion. > > What do you mean by bumping INC_PR? I mean increasing the number of INC_PR to trigger a new build of the pack= age and every dependent package. > After applying your patch, is there anything else I have to consider, l= ike > e.g. cleaning meta-toolchain or things like that? If you increase the PR everything should work without cleaning. Regards, Stefan