Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753710Ab1DUOUA (ORCPT ); Thu, 21 Apr 2011 10:20:00 -0400 Received: from mtagate5.uk.ibm.com ([194.196.100.165]:54609 "EHLO mtagate5.uk.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752733Ab1DUOT7 (ORCPT ); Thu, 21 Apr 2011 10:19:59 -0400 Date: Thu, 21 Apr 2011 16:19:49 +0200 From: Jan Glauber To: Rusty Russell Cc: Christoph Hellwig , linux-kernel@vger.kernel.org, castet.matthieu@free.fr, sliakh.lkml@gmail.com, jiang@cs.ncsu.edu, mingo@elte.hu Subject: Re: Undoing module RONX protection Message-ID: <20110421141949.GA10675@hal> References: <20110418092348.GB7786@hal> <20110418092801.GC3837@infradead.org> <874o5v6drb.fsf@rustcorp.com.au> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <874o5v6drb.fsf@rustcorp.com.au> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2897 Lines: 67 On Mon, Apr 18, 2011 at 08:13:36PM +0930, Rusty Russell wrote: > On Mon, 18 Apr 2011 11:23:48 +0200, Jan Glauber wrote: > > While debugging I stumbled over two problems in the code that protects module > > pages. > > > > First issue is that disabling the protection before freeing init or unload of > > a module is not symmetric with the enablement. For instance, if pages are set > > to RO the page range from module_core to module_core + core_ro_size is > > protected. If a module is unloaded the page range from module_core to > > module_core + core_size is set back to RW. > > So pages that were not set to RO are also changed to RW. > > This is not critical but IMHO it should be symmetric. > > > > Second issue is that while set_memory_rw & set_memory_ro are used for > > RO/RW changes only set_memory_nx is involved for NX/X. One would await that > > the inverse function is called when the NX protection should be removed, > > which is not the case here, unless I'm missing something. > > > > The following patch addresses both issues. Works on s390. Boot tested on x86. > > > > Please comment, > > Applied, minus the S/390 EXPORT_SYMBOL which Christoph pointed out. I > turned your mail into the commit message, since it was clearer and more > verbose. I don't see why they would be different. There's a bug in my patch which just killed one of my s390 machines. Can you merge this with the previuos patch? thanks, Jan --- From: Jan Glauber A module may not have an init part so set_page_attributes may try to change an invalid page range. Modifying page 0 is fatal on s390. Bail out in case the module_core or module_init is not present. Signed-off-by: Jan Glauber --- kernel/module.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) --- a/kernel/module.c +++ b/kernel/module.c @@ -1610,14 +1610,14 @@ static void set_section_ro_nx(void *base /* Setting memory back to W+X before releasing it */ void unset_section_ro_nx(struct module *mod, void *module_region) { - if (mod->module_core == module_region) { + if (mod->module_core && mod->module_core == module_region) { set_page_attributes(mod->module_core + mod->core_text_size, mod->module_core + mod->core_size, set_memory_x); set_page_attributes(mod->module_core, mod->module_core + mod->core_ro_size, set_memory_rw); - } else if (mod->module_init == module_region) { + } else if (mod->module_init && mod->module_init == module_region) { set_page_attributes(mod->module_init + mod->init_text_size, mod->module_init + mod->init_size, set_memory_x); -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/