From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andreas Enge Subject: Re: libgd security update / i686 issues Date: Thu, 28 Jul 2016 10:34:20 +0200 Message-ID: <20160728083420.GA1693@solar> References: <20160728072337.GA1011@jasmine> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:41757) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bSglx-0002CC-NP for guix-devel@gnu.org; Thu, 28 Jul 2016 04:34:30 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bSglt-00048M-I3 for guix-devel@gnu.org; Thu, 28 Jul 2016 04:34:28 -0400 Received: from mailrelay1.public.one.com ([91.198.169.124]:64203) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bSglt-000489-5l for guix-devel@gnu.org; Thu, 28 Jul 2016 04:34:25 -0400 Content-Disposition: inline In-Reply-To: <20160728072337.GA1011@jasmine> List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: "Guix-devel" To: Leo Famulari Cc: guix-devel@gnu.org On Thu, Jul 28, 2016 at 03:23:37AM -0400, Leo Famulari wrote: > Should these CFLAGS values be applied unconditionally, as in the > attached patch, or should they be applied only while building on or for > specific architectures? Or something else? They only work on x86 processors, almost by definition: SSE stands for a certain instruction set. So one would need to check whether the problem occurs for other architectures. I would assume that it happens on all 32 bit architectures, in particular armhf. Their code is too fragile: One should not rely on fine details of the processor architecture or instruction set to hope for an expected rounding behaviour. Andreas