From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932383AbcBHR6R (ORCPT ); Mon, 8 Feb 2016 12:58:17 -0500 Received: from mx1.redhat.com ([209.132.183.28]:47408 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932277AbcBHR6N (ORCPT ); Mon, 8 Feb 2016 12:58:13 -0500 Date: Mon, 8 Feb 2016 12:58:10 -0500 From: Jessica Yu To: Josh Poimboeuf Cc: Seth Jennings , Jiri Kosina , Vojtech Pavlik , Miroslav Benes , Rusty Russell , Steven Rostedt , Ingo Molnar , live-patching@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: Fix ordering of ftrace/livepatch calls on module load and unload Message-ID: <20160208175809.GA27290@packer-debian-8-amd64.digitalocean.com> References: <1454728097-7106-1-git-send-email-jeyu@redhat.com> <20160208174846.GB23106@treble.redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Disposition: inline In-Reply-To: <20160208174846.GB23106@treble.redhat.com> X-OS: Linux eisen.io 3.16.0-4-amd64 x86_64 User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org +++ Josh Poimboeuf [08/02/16 11:48 -0600]: >On Fri, Feb 05, 2016 at 10:08:15PM -0500, Jessica Yu wrote: >> As explained here [1], livepatch modules are failing to initialize properly >> because the ftrace coming module notifier (which calls >> ftrace_module_enable()) runs *after* the livepatch module notifier (which >> enables the patch(es)). Thus livepatch attempts to apply patches to >> modules before ftrace_module_enable() is even called for the corresponding >> module(s). As a result, patch modules break. Ftrace code must run before >> livepatch on module load, and the reverse is true on module unload. >> >> For ftrace and livepatch, order of initialization (plus exit/cleanup code) is >> important for loading and unloading modules, and using module notifiers to >> perform this work is not ideal since it is not always clear what gets called >> when. In this patchset, dependence on the module notifier call chain is removed >> in favor of hard coding the corresponding function calls in the module loader. >> This promotes better code visibility and ensures that ftrace and livepatch code >> get called in the correct order on patch module load and unload. >> >> Tested the changes with a test livepatch module that patches 9p and nilfs2, >> and verified that the issue described in [1] is fixed. > >Since Rusty agreed to your suggested changes for splitting up >complete_formation() and for setting mod->state = MODULE_STATE_GOING >before calling the going notifiers in the error path, can you do a v4 >with those changes? They should probably be split up like: > >1. split up complete_formation() >2. set MODULE_STATE_GOING before calling going notifiers in error path >3. remove ftrace module notifier >4. remove livepatch module notifier Sure, I'll do that. Jessica