From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751119AbbG3ETn (ORCPT ); Thu, 30 Jul 2015 00:19:43 -0400 Received: from mail.skyhub.de ([78.46.96.112]:52056 "EHLO mail.skyhub.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750732AbbG3ETl (ORCPT ); Thu, 30 Jul 2015 00:19:41 -0400 Date: Thu, 30 Jul 2015 06:19:33 +0200 From: Borislav Petkov To: Andy Lutomirski Cc: Andy Lutomirski , X86 ML , "linux-kernel@vger.kernel.org" , Brian Gerst , Steven Rostedt , Willy Tarreau , Thomas Gleixner , Peter Zijlstra , Linus Torvalds Subject: Re: [PATCH v2 3/3] x86/entry/64: Move #BP from IST to the IRQ stack Message-ID: <20150730041933.GB7209@nazgul.tnic> References: <4cdc8c9449c8439d1316e0b56f34870a5e348d73.1437803711.git.luto@kernel.org> <20150728095402.GE1612@nazgul.tnic> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: 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 On Wed, Jul 29, 2015 at 10:57:26AM -0700, Andy Lutomirski wrote: > OK if I do that as a follow-up? It would probably want to be a > separate patch anyway. Of course. > Hmm, I'm starting to like this new regime in which we never ever > switch to user mode from anywhere other than the standard kernel > stack. It looks like even Xen may play along and do it cleanly soon > :) Maybe I'll even add an assertion somewhere to make sure we don't > break it. (I think this also means that the bad iret fixup can be > simplified.) Definitely sounds like a nice, logical thing. We sometimes switch stacks to land on the kernel stack before returning to user mode (IST and all) but I guess that's a clean enough thing to do. Oh, and only a couple of insns so yeah. > Also, with all this stuff applied (and the modify_ldt thing, once the > Xen folks figure out what's wrong), I think we can reinstate the old > LARL check for 16-bit segments and thus prevent naughty users from > banging on espfix using only sigreturn. Uuh, and then only check ZF. I guess this should cover all the legacy cases, which is nice. Yeah, sounds coolio. :-) -- Regards/Gruss, Boris. ECO tip #101: Trim your mails when you reply. --