From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH 5/5] Nested VMX patch 5 implements vmlaunch and vmresume Date: Thu, 22 Oct 2009 12:55:38 +0200 Message-ID: <4AE03A2A.30503@redhat.com> References: <1255617706-13564-1-git-send-email-oritw@il.ibm.com> <1255617706-13564-2-git-send-email-oritw@il.ibm.com> <1255617706-13564-3-git-send-email-oritw@il.ibm.com> <1255617706-13564-4-git-send-email-oritw@il.ibm.com> <1255617706-13564-5-git-send-email-oritw@il.ibm.com> <1255617706-13564-6-git-send-email-oritw@il.ibm.com> <20091019172939.GH29477@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Gleb Natapov , Abel Gordon , aliguori@us.ibm.com, Ben-Ami Yassour1 , kvm@vger.kernel.org, mdday@us.ibm.com, Muli Ben-Yehuda To: Orit Wasserman Return-path: Received: from mx1.redhat.com ([209.132.183.28]:65291 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754800AbZJVKzt (ORCPT ); Thu, 22 Oct 2009 06:55:49 -0400 In-Reply-To: Sender: kvm-owner@vger.kernel.org List-ID: On 10/21/2009 04:43 PM, Orit Wasserman wrote: > > It is possible L1 called vmlaunch but we didn't actually run L2 (for > example there was an interrupt and > enable_irq_window switched back to L1 before running L2). L1 thinks the > vmlaunch was successful and call vmresume in the next time > but KVM needs to call vmlaunch for L2. > Is it really possible? If vmlaunch is started, it should complete unconditionally for L1. The irq window should be recalculated based on the guest vmcs "exit on external interrupt" and guest eflags/cr8 etc. -- error compiling committee.c: too many arguments to function