Return-Path: Received: from mail-qw0-f46.google.com ([209.85.216.46]:60721 "EHLO mail-qw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932383Ab0KWBPZ convert rfc822-to-8bit (ORCPT ); Mon, 22 Nov 2010 20:15:25 -0500 Received: by qwc9 with SMTP id 9so1495431qwc.19 for ; Mon, 22 Nov 2010 17:15:24 -0800 (PST) From: Ben Gamari Subject: Kernel build under meta-toolchain SDK fails Date: Mon, 22 Nov 2010 20:15:20 -0500 Message-ID: <87y68kx2l3.fsf@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8BIT Sender: linux-kbuild-owner@vger.kernel.org List-ID: To: openembedded-devel@lists.openembedded.org Cc: =?utf-8?Q?Am=C3=A9rico?= Wang , Christoph Egger , Sam Ravnborg , linux-kbuild Hey all, Recently I tried updating our openembedded environment for an embedded application built around the BeagleBoard. Unfortunately, the build failed while configuring the kernel with an inexplicable bus error. To trace down the issue I installed the latest meta-toolchain sdk and tried building a kernel. This resulted in an identical failure, $ make HOSTCC scripts/basic/fixdep In file included from /usr/local/angstrom/arm/arm-angstrom-linux-gnueabi/usr/include/stdlib.h:903, from scripts/basic/fixdep.c:112: /usr/local/angstrom/arm/arm-angstrom-linux-gnueabi/usr/include/bits/stdlib.h:65: warning: no previous prototype for ‘ptsname_r’ /bin/sh: line 1: 4192 Bus error scripts/basic/fixdep scripts/basic/.fixdep.d scripts/basic/fixdep 'gcc -Wp,-MD,scripts/basic/.fixdep.d -Wall -Wmissing-prototypes -Wstrict-prototypes -O2 -fomit-frame-pointer -o scripts/basic/fixdep scripts/basic/fixdep.c ' > scripts/basic/.fixdep.tmp make[2]: *** [scripts/basic/fixdep] Error 135 make[1]: *** [scripts_basic] Error 2 make: *** No rule to make target `include/config/auto.conf', needed by `include/config/kernel.release'. Stop. The problem is quite early in the build since even make oldconfig fails in this way. I tried tracking down the bug on the LKML[1] but after some discussion it seems very likely the problem is a toolchain issue. At least part of the problem seems to be in fixdep, which fails very strangely in fstat[2]. I've done quite a bit of debugging[2], but it seems to be a bit of a heisenbug. Interestingly enough, I can reproduce the issue both on my personal laptop (running Ubuntu 10.10, gcc 4.4.5) and our build server (running Ubuntu 10.04, gcc 4.4.3). It would be great if someone could try building a kernel from master using openembedded master. Otherwise, any ideas would be greatly appreciated. Cheers, - Ben [1] http://lkml.org/lkml/2010/11/6/90 [2] http://lkml.org/lkml/2010/11/7/78