From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934077AbdDGTZ6 (ORCPT ); Fri, 7 Apr 2017 15:25:58 -0400 Received: from Galois.linutronix.de ([146.0.238.70]:44530 "EHLO Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754895AbdDGTZv (ORCPT ); Fri, 7 Apr 2017 15:25:51 -0400 Date: Fri, 7 Apr 2017 21:25:30 +0200 (CEST) From: Thomas Gleixner To: Mathias Krause cc: Andy Lutomirski , Kees Cook , Andy Lutomirski , "kernel-hardening@lists.openwall.com" , Mark Rutland , Hoeun Ryu , PaX Team , Emese Revfy , Russell King , X86 ML , "linux-kernel@vger.kernel.org" , "linux-arm-kernel@lists.infradead.org" , Peter Zijlstra Subject: Re: [kernel-hardening] Re: [RFC v2][PATCH 04/11] x86: Implement __arch_rare_write_begin/unmap() In-Reply-To: Message-ID: References: <1490811363-93944-1-git-send-email-keescook@chromium.org> <1490811363-93944-5-git-send-email-keescook@chromium.org> User-Agent: Alpine 2.20 (DEB 67 2015-01-07) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 7 Apr 2017, Mathias Krause wrote: > On 7 April 2017 at 15:14, Thomas Gleixner wrote: > > On Fri, 7 Apr 2017, Mathias Krause wrote: > >> On 7 April 2017 at 11:46, Thomas Gleixner wrote: > >> > Whether protected by preempt_disable or local_irq_disable, to make that > >> > work it needs CR0 handling in the exception entry/exit at the lowest > >> > level. And that's just a nightmare maintainence wise as it's prone to be > >> > broken over time. > >> > >> It seems to be working fine for more than a decade now in PaX. So it > >> can't be such a big maintenance nightmare ;) > > > > I really do not care whether PaX wants to chase and verify that over and > > over. I certainly don't want to take the chance to leak CR0.WP ever and I > > very much care about extra stuff to check in the entry/exit path. > > Fair enough. However, placing a BUG_ON(!(read_cr0() & X86_CR0_WP)) > somewhere sensible should make those "leaks" visible fast -- and their > exploitation impossible, i.e. fail hard. Sure, you trade leaking WP with an potentially exploitable BUG(). > >> The "proper solution" seems to be much slower compared to just > >> toggling CR0.WP (which is costly in itself, already) because of the > >> TLB invalidation / synchronisation involved. > > > > Why the heck should we care about rare writes being performant? > > As soon as they stop being rare and people start extending the r/o > protection to critical data structures accessed often. Then > performance matters. Emphasis on "Then". I'm not seeing it, because no matter what you do it's going to be slow. Aside of that increasing the usage will also increase the chance to leak stuff. In that case I rather leak a single page mapping temporarily than taking the chance to leak WP. > >> > It's valid (at least on x86) to have a shadow map with the same page > >> > attributes but write enabled. That does not require any fixups of CR0 and > >> > just works. > >> > >> "Just works", sure -- but it's not as tightly focused as the PaX > >> solution which is CPU local, while your proposed solution is globally > >> visible. > > > > Making the world and some more writeable hardly qualifies as tightly > > focussed. Making the mapping concept CPU local is not rocket science > > either. The question is whethers it's worth the trouble. > > No, the question is if the value of the concept is well understood and > if people can see what could be done with such a strong primitive. > Apparently not... Oh, well. We can stop that discussion right here, if all you can provide is a killer phrase. I'm well aware what can be done with a strong primitive and I certainly understand the concept, but I'm not naive enough to believe that lifting one of the strong protections the kernel has by globaly disabling WP is anything which should be even considered. That bit is a horrible misconception and should be fused to 1. Aside of that, if you had taken the time to figure out how kmap_atomic stuff works then you would have noticed that it does not require cross CPU pagetable syncs and that the mapping place can be randomized to a certain degree. So this has neither global impact, nor does it become immediately globally visible. Thanks, tglx