From mboxrd@z Thu Jan 1 00:00:00 1970 From: "H. Peter Anvin" Subject: Re: [PATCH] Revert "x86-64, init: Do not set NX bits on non-NX capable hardware" Date: Mon, 20 May 2013 11:19:11 -0700 Message-ID: <519A691F.9060007@linux.intel.com> References: <20130514160108.GA5055@dhcp22.suse.cz> <1369062192-7066-1-git-send-email-mhocko@suse.cz> <519A483F.7070400@zytor.com> <20130520164335.GA9441@dhcp22.suse.cz> <519A67C9.6000709@linux.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: Received: from mga14.intel.com ([143.182.124.37]:26830 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757796Ab3ETSTM (ORCPT ); Mon, 20 May 2013 14:19:12 -0400 In-Reply-To: Sender: linux-pm-owner@vger.kernel.org List-Id: linux-pm@vger.kernel.org To: Linus Torvalds Cc: Michal Hocko , "H. Peter Anvin" , =?UTF-8?B?RmVybmFuZG8gTHVpcyBWw6F6cXVleiBDYW8=?= , "Rafael J. Wysocki" , the arch/x86 maintainers , "linux-pm@vger.kernel.org" , Linux Kernel Mailing List On 05/20/2013 11:18 AM, Linus Torvalds wrote: > On Mon, May 20, 2013 at 11:13 AM, H. Peter Anvin wrote: >> >> Indeed. Looks like we corrupt a random bit in memory on resume. > > Well, in this case, it's probably the DEBUG_PAGEALLOC that Michal has > enabled, which causes a page fault - and then a triple fault - at > secondary CPU startup. And the triple fault then results in a reboot. > Which explains the symptoms he gets. > > Without DEBUG_PAGEALLOC it ends up being "just" a single-bit corruption, yes. > Good, that explains what made it predictable. Score one for DEBUG_PAGEALLOC. -hpa