From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [BUG] kvm crashes in 2.6.28-rc6-00007-ged31348 Date: Tue, 02 Dec 2008 14:23:52 +0200 Message-ID: <493528D8.8010904@redhat.com> References: <20081130193401.GA7690@hades> <4932F59D.2060002@redhat.com> <20081130203833.GA14903@hades> <4932FB47.6090300@redhat.com> <20081130210406.GA17952@hades> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org To: Luis Henriques Return-path: Received: from mx2.redhat.com ([66.187.237.31]:34000 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753244AbYLBMX6 (ORCPT ); Tue, 2 Dec 2008 07:23:58 -0500 In-Reply-To: <20081130210406.GA17952@hades> Sender: kvm-owner@vger.kernel.org List-ID: Luis Henriques wrote: > On Sun, Nov 30, 2008 at 10:44:55PM +0200, Avi Kivity wrote: > >> Luis Henriques wrote: >> >>> No, I was not able to reproduce the issue. Please let me know if you need some >>> more information on my system (.config, for instance). >>> >>> >> Were you using some other virtualization product? Were you running >> suspend/resume? >> > > No for both questions. However, I had compiled support for suspend (not sure if > this is what you mean by "running suspend/resume") - This is a feature I used > only once or twice... > The underlying problem is that an svm instruction has been executed, but svm is disabled. Since kvm enables svm unconditionally on all processors on startup, there are only a few paths that can potentially trigger this: - another virtualization module turned svm off - cpu hotadd/hotremove (suspend/resume triggers this) - something did a read-modify-write cycle on cr4 (which contains the svm enable bit) while kvm enabled that bit - core was turned off (does linux power management do that?) Anything ring a bell? -- error compiling committee.c: too many arguments to function