From mboxrd@z Thu Jan 1 00:00:00 1970 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754419AbeAJL1V (ORCPT + 1 other); Wed, 10 Jan 2018 06:27:21 -0500 Received: from mx1.redhat.com ([209.132.183.28]:55770 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754217AbeAJL1T (ORCPT ); Wed, 10 Jan 2018 06:27:19 -0500 Subject: Re: [PATCH v3 3/5] x86/enter: Use IBRS on syscall and interrupts To: Peter Zijlstra , Tim Chen Cc: Thomas Gleixner , Andy Lutomirski , Linus Torvalds , Greg KH , Dave Hansen , Andrea Arcangeli , Andi Kleen , Arjan Van De Ven , David Woodhouse , Dan Williams , Ashok Raj , linux-kernel@vger.kernel.org References: <20180110100457.GA29822@worktop.programming.kicks-ass.net> From: Paolo Bonzini Message-ID: <9e8f46f6-35a7-e38d-0197-fb86b40fde1a@redhat.com> Date: Wed, 10 Jan 2018 12:27:09 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.5.0 MIME-Version: 1.0 In-Reply-To: <20180110100457.GA29822@worktop.programming.kicks-ass.net> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.30]); Wed, 10 Jan 2018 11:27:19 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Return-Path: On 10/01/2018 11:04, Peter Zijlstra wrote: > On Tue, Jan 09, 2018 at 06:26:47PM -0800, Tim Chen wrote: >> Set IBRS upon kernel entrance via syscall and interrupts. Clear it >> upon exit. IBRS protects against unsafe indirect branching predictions >> in the kernel. >> >> The NMI interrupt save/restore of IBRS state was based on Andrea >> Arcangeli's implementation. >> Here's an explanation by Dave Hansen on why we save IBRS state for NMI. >> >> The normal interrupt code uses the 'error_entry' path which uses the >> Code Segment (CS) of the instruction that was interrupted to tell >> whether it interrupted the kernel or userspace and thus has to switch >> IBRS, or leave it alone. >> >> The NMI code is different. It uses 'paranoid_entry' because it can >> interrupt the kernel while it is running with a userspace IBRS (and %GS >> and CR3) value, but has a kernel CS. If we used the same approach as >> the normal interrupt code, we might do the following; >> >> SYSENTER_entry >> <-------------- NMI HERE >> IBRS=1 >> do_something() >> IBRS=0 >> SYSRET >> >> The NMI code might notice that we are running in the kernel and decide >> that it is OK to skip the IBRS=1. This would leave it running >> unprotected with IBRS=0, which is bad. >> >> However, if we unconditionally set IBRS=1, in the NMI, we might get the >> following case: >> >> SYSENTER_entry >> IBRS=1 >> do_something() >> IBRS=0 >> <-------------- NMI HERE (set IBRS=1) >> SYSRET >> >> and we would return to userspace with IBRS=1. Userspace would run >> slowly until we entered and exited the kernel again. >> >> Instead of those two approaches, we chose a third one where we simply >> save the IBRS value in a scratch register (%r13) and then restore that >> value, verbatim. >> > > What this Changelog fails to address is _WHY_ we need this. What does > this provide that retpoline does not. Which, for the record, is just that it works better on Skylake+ CPUs. Paolo