From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bruce MaJia Subject: Re: question on KVM run flow on Intel-VT Date: Wed, 22 Apr 2009 21:24:11 +0800 Message-ID: <20090422132411.GA10001@LL.wrs.com> References: <20090422125549.GA8186@LL> Reply-To: Bruce MaJia Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii To: kvm@vger.kernel.org Return-path: Received: from rv-out-0506.google.com ([209.85.198.229]:5881 "EHLO rv-out-0506.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752652AbZDVNYV (ORCPT ); Wed, 22 Apr 2009 09:24:21 -0400 Received: by rv-out-0506.google.com with SMTP id f9so2898614rvb.1 for ; Wed, 22 Apr 2009 06:24:20 -0700 (PDT) Content-Disposition: inline In-Reply-To: <20090422125549.GA8186@LL> Sender: kvm-owner@vger.kernel.org List-ID: On Wed, Apr 22, 2009 at 08:55:49PM +0800, MaJia wrote: > Hi, > > I am learning kvm + Intel-VT nowadays. Since I am rookie on both kvm and > Intel-VT, I got following questions about run flow of kvm on vmx. > > I think the cpu is running in 'vmx non-root' mode when execution flow > reach 'vmx_vcpu_run' and the batch of 'asm code' take responsibility of > run guest on cpu. > > My question is, when and where will the execution flow return to vmx root > mode? I may not clarify my question. I mean, when vmexit occur, which caused by executing io instruction from guest, will the cpu switch to vmx root mode? Will the cpu direct goto run 'handle_io' function? When cpu running in vmx non-root mode, an interrupt occurs, what is the behavior of cpu? Thanks. Bruce. > > Please give me some hint about this point or route me to some documents. > > Thanks. > Bruce.