From owner-freebsd-ports-bugs@FreeBSD.ORG Mon Apr 27 13:20:02 2009 Return-Path: Delivered-To: freebsd-ports-bugs@hub.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D356B106564A for ; Mon, 27 Apr 2009 13:20:02 +0000 (UTC) (envelope-from gnats@FreeBSD.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:4f8:fff6::28]) by mx1.freebsd.org (Postfix) with ESMTP id C12F58FC15 for ; Mon, 27 Apr 2009 13:20:02 +0000 (UTC) (envelope-from gnats@FreeBSD.org) Received: from freefall.freebsd.org (gnats@localhost [127.0.0.1]) by freefall.freebsd.org (8.14.3/8.14.3) with ESMTP id n3RDK2vb082563 for ; Mon, 27 Apr 2009 13:20:02 GMT (envelope-from gnats@freefall.freebsd.org) Received: (from gnats@localhost) by freefall.freebsd.org (8.14.3/8.14.3/Submit) id n3RDK20K082562; Mon, 27 Apr 2009 13:20:02 GMT (envelope-from gnats) Date: Mon, 27 Apr 2009 13:20:02 GMT Message-Id: <200904271320.n3RDK20K082562@freefall.freebsd.org> To: freebsd-ports-bugs@FreeBSD.org From: Alexander Churanov Cc: Subject: Re: ports/133487: endianess detection wrong in devel/boost X-BeenThere: freebsd-ports-bugs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: Alexander Churanov List-Id: Ports bug reports List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 27 Apr 2009 13:20:03 -0000 The following reply was made to PR ports/133487; it has been noted by GNATS. From: Alexander Churanov To: bug-followup@freebsd.org Cc: Subject: Re: ports/133487: endianess detection wrong in devel/boost Date: Mon, 27 Apr 2009 17:19:21 +0400 Resolution: Suggest applying this patch and keeping it until updating to boost-1.38. Analysis: The root cause of the issue is boost matching endianess using method of GNU libc, which is not compatible with the way BSDs define endianess. They've fixed this in 1.38 by applying a patch from NetBSD. See https://svn.boost.org/trac/boost/ticket/1922 for details. Future directions: The fix applied in 1.38 is to match against CPU architecture if matching using GNU method fails. This may re-introduce bugs for platforms, not specifically listed in boost header file. Suggest proposing the patch, attached to this issue, for permanent inclusion into boost libraries. Alexander Churanov, maintainer of devel/boost