From owner-freebsd-arm@freebsd.org Mon Jun 13 18:38:01 2016 Return-Path: Delivered-To: freebsd-arm@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 0CF83AF22A6 for ; Mon, 13 Jun 2016 18:38:01 +0000 (UTC) (envelope-from lausts@laus.org) Received: from cdptpa-oedge-vip.email.rr.com (cdptpa-outbound-snat.email.rr.com [107.14.166.228]) by mx1.freebsd.org (Postfix) with ESMTP id C46262F65 for ; Mon, 13 Jun 2016 18:38:00 +0000 (UTC) (envelope-from lausts@laus.org) Received: from [173.88.10.122] ([173.88.10.122:26085] helo=mail.laus.org) by cdptpa-oedge01 (envelope-from ) (ecelerity 3.5.0.35861 r(Momo-dev:tip)) with ESMTP id 2E/95-12840-BE9FE575; Mon, 13 Jun 2016 18:22:35 +0000 Received: from mail.laus.org (localhost [127.0.0.1]) by mail.laus.org (8.15.2/8.15.2) with ESMTPS id u5DIMYEO072313 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO) for ; Mon, 13 Jun 2016 14:22:34 -0400 (EDT) (envelope-from lausts@laus.org) Received: (from lausts@localhost) by mail.laus.org (8.15.2/8.15.2/Submit) id u5DIMY1s072312 for freebsd-arm@freebsd.org; Mon, 13 Jun 2016 14:22:34 -0400 (EDT) (envelope-from lausts) Date: Mon, 13 Jun 2016 14:22:34 -0400 From: Thomas Laus To: freebsd-arm@freebsd.org Subject: Crossbuild Failure on Arm Message-ID: <20160613182234.GA72262@mail.laus.org> Reply-To: lausts@acm.org MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline X-Operating-System: FreeBSD 10.3-RELEASE-p5 on an amd64 User-Agent: Mutt/1.6.1 (2016-04-27) X-RR-Connecting-IP: 107.14.168.118:25 X-Cloudmark-Score: 0 X-BeenThere: freebsd-arm@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: "Porting FreeBSD to ARM processors." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Jun 2016 18:38:01 -0000 I wanted to add a few NIC cards to my Beaglebone, so my next step was to build a toolchain for the arm arch. I used the instructions from the khubla.com site which starts with a build of the necessary toolchain with the command: make XDEV=arm XDEV_ARCH=armv6 xdev It failed! sh /usr/src/tools/install.sh -C -o root -g wheel -m 444 /usr/src/gnu/lib/libdialog/../../../contrib/dialog/dialog.h /usr/src/gnu/lib/libdialog/../../../contrib/dialog/dlg_colors.h /usr/src/gnu/lib/libdialog/dlg_config.h /usr/src/gnu/lib/libdialog/../../../contrib/dialog/dlg_keys.h //usr/armv6-freebsd/usr/include/ ===> lib/libc++ (obj,all,install) /usr/obj/armv6-freebsd/usr/src/lib/libc++ created for /usr/src/lib/libc++ echo libc++.so.1.full: //usr/armv6-freebsd/usr/lib/libcxxrt.a >> .depend c++ -isystem //usr/armv6-freebsd/usr/include -L//usr/armv6-freebsd/usr/lib --sysroot=//usr/armv6-freebsd/ -B//usr/armv6-freebsd/usr/libexec -B//usr/armv6-freebsd/usr/bin -B//usr/armv6-freebsd/usr/lib -mlong-calls -O -pipe -isystem /usr/src/lib/libc++/../../contrib/libc++/include -isystem /usr/src/lib/libc++/../../contrib/libcxxrt -nostdinc++ -nostdlib -DLIBCXXRT -MD -MF.depend.algorithm.o -MTalgorithm.o -Qunused-arguments -std=c++11 -Wno-c++11-extensions -c /usr/src/lib/libc++/../../contrib/libc++/src/algorithm.cpp -o algorithm.o In file included from /usr/src/lib/libc++/../../contrib/libc++/src/algorithm.cpp:10: In file included from /usr/src/lib/libc++/../../contrib/libc++/include/algorithm:624: In file included from /usr/src/lib/libc++/../../contrib/libc++/include/initializer_list:47: /usr/src/lib/libc++/../../contrib/libc++/include/cstddef:43:15: fatal error: 'stddef.h' file not found #include_next ^ 1 error generated. *** Error code 1 Stop. make[4]: stopped in /usr/src/lib/libc++ *** Error code 1 Stop. make[3]: stopped in /usr/src *** Error code 1 Stop. make[2]: stopped in /usr/src *** Error code 1 Stop. make[1]: stopped in /usr/src *** Error code 1 Stop. make: stopped in /usr/src I have made a fresh checkout of the HEAD sources about every three days and make another attempt. The operation faults at exactly the same point each time. Am I doing something wrong or is there a bug in Clang? The same checkout builds normally in native amd64 on an Intel i5 PC. I thought that the Jenkins build would have caught any Clang bug. Do I need to file a bug report? Tom -- Public Keys: PGP KeyID = 0x5F22FDC1 GnuPG KeyID = 0x620836CF