From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com ([66.187.233.31]:55897 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752426AbXFOMKr (ORCPT ); Fri, 15 Jun 2007 08:10:47 -0400 From: David Howells In-Reply-To: <200706151355.57464.ak@suse.de> References: <200706151355.57464.ak@suse.de> <200706150159.l5F1xNgM000459@hera.kernel.org> <200706151128.39566.arnd@arndb.de> <200706151131.38429.arnd@arndb.de> Subject: Re: [PATCH] Introduce compat_u64 and compat_s64 types Date: Fri, 15 Jun 2007 13:09:17 +0100 Message-ID: <5717.1181909357@redhat.com> Sender: linux-arch-owner@vger.kernel.org To: Andi Kleen Cc: Arnd Bergmann , Benjamin Herrenschmidt , David Woodhouse , Linux Kernel Mailing List , Dave Airlie , linux-arch@vger.kernel.org, Andrew Morton List-ID: Andi Kleen wrote: > Why does it break them? It should just make them a little slower. Not all CPUs deliver recoverable misalignment exceptions. This is probably particularly true of NOMMU-mode archs where the CPU designed may have taken the view that if a data exception is delivered, then the whole system is kaput anyway and must be restarted. > The network code requires unaligned accesses to work anyways so if your > architecture doesn't support them it is already remotely crashable. I thought we'd fixed all that. David