From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757564Ab2BBWHv (ORCPT ); Thu, 2 Feb 2012 17:07:51 -0500 Received: from terminus.zytor.com ([198.137.202.10]:52026 "EHLO mail.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751322Ab2BBWHu (ORCPT ); Thu, 2 Feb 2012 17:07:50 -0500 Message-ID: <4F2B090F.5020609@zytor.com> Date: Thu, 02 Feb 2012 14:07:11 -0800 From: "H. Peter Anvin" User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:9.0) Gecko/20111222 Thunderbird/9.0 MIME-Version: 1.0 To: Steven Rostedt CC: Ingo Molnar , linux-kernel@vger.kernel.org, Andrew Morton , Jason Baron , Frederic Weisbecker Subject: Re: [PATCH 0/5 v2] [GIT PULL] x86/jump label: Paranoid checks and 2 or 5 byte nops References: <20120128160925.275163704@goodmis.org> <20120130144857.GA22013@elte.hu> <1327936060.22710.165.camel@gandalf.stny.rr.com> <20120131194813.GA12074@elte.hu> <1328072925.5882.57.camel@gandalf.stny.rr.com> <20120201080539.GA11770@elte.hu> <1328109697.5882.65.camel@gandalf.stny.rr.com> <20120202135154.GA10315@elte.hu> <4F2A9960.2090500@zytor.com> <1328193068.5882.95.camel@gandalf.stny.rr.com> <1328220280.5882.122.camel@gandalf.stny.rr.com> In-Reply-To: <1328220280.5882.122.camel@gandalf.stny.rr.com> Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 02/02/2012 02:04 PM, Steven Rostedt wrote: > commit 7bbd9688727dd30cd44b90b3c9c35f0f66af77ff > Author: Steven Rostedt > Date: Thu Feb 2 16:55:43 2012 -0500 > > x86/jump labels: Handle initialization of enabled nops > > When jump labels are initialized at boot up, they are compared > to the default_nop before switching to the ideal nop. > > But if a jump label is enabled by default on start up, the > enabled code does not test against the default nop, only the > ideal nop. But as this jump label has not been converted to the > ideal nop, it fails the check, and will crash the box. > > The enabled path needs to be aware of initialization too. > > Reported-by: Ingo Molnar > Signed-off-by: Steven Rostedt > Would it make more sense to keep track of three states: "Default" "Enabled" "Disabled" -?