From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 17725 invoked by alias); 22 Jun 2009 23:02:48 -0000 Received: (qmail 17602 invoked by uid 22791); 22 Jun 2009 23:02:48 -0000 X-SWARE-Spam-Status: No, hits=-2.4 required=5.0 tests=AWL,BAYES_00,SPF_PASS X-Spam-Check-By: sourceware.org Received: from mail-ew0-f228.google.com (HELO mail-ew0-f228.google.com) (209.85.219.228) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Mon, 22 Jun 2009 23:02:42 +0000 Received: by ewy28 with SMTP id 28so5197546ewy.24 for ; Mon, 22 Jun 2009 16:02:39 -0700 (PDT) Received: by 10.210.91.7 with SMTP id o7mr5571451ebb.48.1245711759017; Mon, 22 Jun 2009 16:02:39 -0700 (PDT) Received: from ?192.168.2.99? (cpc2-cmbg8-0-0-cust61.cmbg.cable.ntl.com [82.6.108.62]) by mx.google.com with ESMTPS id 5sm179786eyh.40.2009.06.22.16.02.38 (version=SSLv3 cipher=RC4-MD5); Mon, 22 Jun 2009 16:02:38 -0700 (PDT) Message-ID: <4A401076.5030200@gmail.com> Date: Mon, 22 Jun 2009 23:02:00 -0000 From: Dave Korn User-Agent: Thunderbird 2.0.0.17 (Windows/20080914) MIME-Version: 1.0 To: Pierre Muller CC: binutils@sourceware.org, gdb@sourceware.org Subject: Re: [Build failure] Cygwin GDB multibuild failure References: <000001c9f389$d74ac340$85e049c0$@u-strasbg.fr> In-Reply-To: <000001c9f389$d74ac340$85e049c0$@u-strasbg.fr> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Mailing-List: contact gdb-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-owner@sourceware.org X-SW-Source: 2009-06/txt/msg00216.txt.bz2 Pierre Muller wrote: > When I try to build GDB > with --enable-targets=all --enable-64bit-bfd > I get the following error: > ../../purecvs/bfd/plugin.c: In function `load_plugin': > ../../purecvs/bfd/plugin.c:206: error: structure has no member named > `d_type' > There is no d_type field for dirent structure on the currently distributed > cygwin. It seems that this field has been added recently > in winsup directory, but is it possible to fix that build failure > nonetheless? As far as I can see, POSIX guarantees nothing beyond the presence of d_ino and d_name[], so use of d_type should probably be protected by some kind of autoconf test. I don't know if there already is a suitable one or not. cheers, DaveK