From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752096AbbASUCp (ORCPT ); Mon, 19 Jan 2015 15:02:45 -0500 Received: from mx1.redhat.com ([209.132.183.28]:40899 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751849AbbASUCo (ORCPT ); Mon, 19 Jan 2015 15:02:44 -0500 Date: Mon, 19 Jan 2015 14:02:39 -0600 From: Josh Poimboeuf To: Jiri Kosina Cc: Seth Jennings , Vojtech Pavlik , live-patching@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] livepatch: support for repatching a function Message-ID: <20150119200239.GA15916@treble.redhat.com> References: <20150119145431.GC1737@treble.redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.23.1-rc1 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Jan 19, 2015 at 08:48:42PM +0100, Jiri Kosina wrote: > On Mon, 19 Jan 2015, Josh Poimboeuf wrote: > > > > If this is implemented really in a fully stackable manner (i.e. you > > > basically would be able to disable only the function that is currently > > > "active", i.e. on top of the stack), woudln't that provide more > > > predictable semantics? > > > > Yes, I agree. Thanks for the comment. > > > > Would you want to enforce stacking even if there are no dependencies > > between the patches? I think that would be easiest (and cleanest). > > Yup, I think that makes the most sense (especially in this "first step"). > Relaxing the revert rules to cover only patches which are really dependent > on each other (and we'd have to be careful about defining the meaning > this, especially with repsect to various consistency models coming in the > future) is something tha can always be done later on top. Sounds good. I'll do a v2. FYI, I've also been working on a prototype of a consistency model, based on my discussions with Vojtech on the list a few months ago (LEAVE_PATCHED_SET + SWITCH_THREAD). I'll probably have some patches to send out for comments in a few weeks. That should hopefully be a good starting point for more discussion about the consistency model(s). -- Josh