From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754850Ab2GIOry (ORCPT ); Mon, 9 Jul 2012 10:47:54 -0400 Received: from mail-pb0-f46.google.com ([209.85.160.46]:65146 "EHLO mail-pb0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754293Ab2GIOru (ORCPT ); Mon, 9 Jul 2012 10:47:50 -0400 Message-ID: <4FFAEF13.7010108@gmail.com> Date: Mon, 09 Jul 2012 08:47:47 -0600 From: David Ahern User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:13.0) Gecko/20120614 Thunderbird/13.0.1 MIME-Version: 1.0 To: Avi Kivity CC: Peter Zijlstra , Gleb Natapov , LKML Subject: Re: perf with precise attribute kills all KVM based VMs References: <4FFAE6D8.5050609@gmail.com> <20120709141904.GC7298@redhat.com> <1341843844.3462.75.camel@twins> <4FFAED16.7050204@redhat.com> In-Reply-To: <4FFAED16.7050204@redhat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 7/9/12 8:39 AM, Avi Kivity wrote: > On 07/09/2012 05:24 PM, Peter Zijlstra wrote: >> On Mon, 2012-07-09 at 17:19 +0300, Gleb Natapov wrote: >>> Yes, this is knows problem that I can't find time to fix. The crash is >>> cause by CPU using host PEBS virtual address while guest is running >>> which causes guest memory corruption. We should disable evens that use >>> PEBS at the guest entry. >> >> Whoops.. so the hardware is reading the DS address as programmed by the >> host (host linear address) and using it as a guest linear address? >> >> Quality stuff.. >> >> Disabling PEBS events for guests isn't pretty though.. > > We already have atomic MSR switching at guest entry/exit time. So it's > not pretty in terms of not getting full profiling, but the code won't be > too hard. Basically we just have to exclude_guest any pebs event. I found this testing changes to perf-kvm, but found the problem extends to just perf-record. With perf-record exclude_guest defaults to 1. See tools/perf/util/util.c, event_attr_init(). > >> but I guess the >> only alternative is mapping the DS into the guest and reprogramming >> MSR_IA32_DS_AREA which is all a little involved I suppose? > > Way too involved, especially as it's virtual addresses and we don't > control the guest cr3. Note that the hardware won't fail gracefully. >