From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from proxy.dresearch.de ([87.193.137.100] helo=mail.dresearch.de) by linuxtogo.org with esmtp (Exim 4.69) (envelope-from ) id 1Oz4Bx-0002hN-Ud for openembedded-devel@lists.openembedded.org; Fri, 24 Sep 2010 11:04:27 +0200 Received: from exchange.intern.dresearch.de (owa.xfer-intern.dresearch.de [192.168.32.16]) by mail.dresearch.de (Postfix) with ESMTP id 545D9491286 for ; Fri, 24 Sep 2010 11:03:37 +0200 (CEST) Received: from [127.0.0.1] ([10.32.10.2]) by exchange.intern.dresearch.de with Microsoft SMTPSVC(6.0.3790.4675); Fri, 24 Sep 2010 11:03:36 +0200 Message-ID: <4C9C696B.4050800@dresearch.de> Date: Fri, 24 Sep 2010 11:03:39 +0200 From: Steffen Sledz User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.9.2.9) Gecko/20100915 Lightning/1.0b2 Thunderbird/3.1.4 MIME-Version: 1.0 To: openembedded-devel@lists.openembedded.org References: <1285241444.3691.60.camel@mattotaupa> <4C9B57F8.6000605@dresearch.de> <4C9B608F.4060602@dresearch.de> In-Reply-To: <4C9B608F.4060602@dresearch.de> X-OriginalArrivalTime: 24 Sep 2010 09:03:36.0899 (UTC) FILETIME=[5F7F5130:01CB5BC7] X-SA-Exim-Connect-IP: 87.193.137.100 X-SA-Exim-Mail-From: sledz@dresearch.de X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on discovery X-Spam-Level: X-Spam-Status: No, score=-2.6 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.2.5 X-SA-Exim-Version: 4.2.1 (built Wed, 25 Jun 2008 17:20:07 +0000) X-SA-Exim-Scanned: Yes (on linuxtogo.org) Subject: Re: libatasmart_0.13.bb: `No package 'libudev' found` and `./strpool: ./strpool: cannot execute binary file` 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: Fri, 24 Sep 2010 09:05:04 -0000 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Am 23.09.2010 16:13, schrieb Steffen Sledz: > On 09/23/2010 03:36 PM, Steffen Sledz wrote: >> Am 23.09.2010 13:30, schrieb Paul Menzel: >>> Using =C8=A6ngstr=C3=B6m 2008.1 this error was not met though. But `d= o_compile()` >>> fails with `./strpool: cannot execute binary file`. >>> >>> Log data follows: >>> | NOTE: make >>> | CC strpool.o >>> | CCLD strpool >>> | ./strpool atasmart.c atasmart.strpool.c >>> | ./strpool: ./strpool: cannot execute binary file >>> | make: *** [atasmart.strpool.c] Error 126 >>> | FATAL: oe_runmake failed >>> | ERROR: Function do_compile failed >>> NOTE: package libatasmart-0.13-r1: task do_compile: Failed >>> >>> But as far as I understand the recipe takes care of that by compiling >>> `strpool` for the build host [2]. >>> >>> do_compile_prepend() { >>> rm strpool -f >>> ${BUILD_CC} -o strpool strpool.c >>> chmod +x strpool >>> } >>> ... >>> [2] http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/d= evicekit/libatasmart_0.13.bb?id=3D67ab615dac83b0460fa1b88dfb11c72d7c93e25= 9 >> >> I hit the same problem. But just when building libatasmart for the fir= st time. A second build succeeds. >=20 > I tried the following to find the cause for this. >=20 > First i added >=20 > pwd > file strpool >=20 > to the do_compile_prepend(). >=20 > Aftre running >=20 > bitbake libatasmart -c clean > bitbake libatasmart >=20 > i saw >=20 > ERROR: log data follows (/home/sledz/work/HydraIP/OE/tmp.6/work/armv5= te-angstrom-linux-gnueabi/libatasmart-0.13-r1/temp/log.do_compile.29576) > | /home/sledz/work/HydraIP/OE/tmp.6/work/armv5te-angstrom-linux-gnuea= bi/libatasmart-0.13-r1/git > | strpool: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), = dynamically linked (uses shared libs), for GNU/Linux 2.6.4, not stripped > ... >=20 > Then i called a devshell >=20 > bitbake libatasmart -c devshell >=20 > Within it >=20 > > pwd > /home/sledz/work/HydraIP/OE/tmp.6/work/armv5te-angstrom-linux-gnueabi= /libatasmart-0.13-r1/git > > file strpool > strpool: ELF 32-bit LSB executable, ARM, version 1 (SYSV), dynamicall= y linked (uses shared libs), for GNU/Linux 2.6.16, not stripped >=20 > So the binary has changed in between. I did some deeper research. The do_compile_prepend code from Koen(?) did = not help here. Makefile.am contains a target "noinst_PROGRAMS =3D strpool" which is used= to create atasmart.strpool.c. For the creation of this target strpool there are no special rules. There= for it is build for the target (and overwrites the strpool built from do_= compile_prepend) instead of the build host. :( Is anybody able to write down a rule for strpool fixing this? Steffen