From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marcelo Tosatti Subject: Re: [patch 0/5] KVM paravirt MMU updates and cr3 caching Date: Sat, 16 Feb 2008 23:24:30 -0300 Message-ID: <20080217022430.GA2454@dmt> References: <20080216220924.733723618@redhat.com>> <47B7739C.8090806@codemonkey.ws> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: kvm-devel@lists.sourceforge.net, Avi Kivity To: Anthony Liguori Return-path: Content-Disposition: inline In-Reply-To: <47B7739C.8090806@codemonkey.ws> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: kvm-devel-bounces@lists.sourceforge.net Errors-To: kvm-devel-bounces@lists.sourceforge.net List-Id: kvm.vger.kernel.org On Sat, Feb 16, 2008 at 05:37:00PM -0600, Anthony Liguori wrote: > Marcelo Tosatti wrote: > >The following patchset, based on earlier work by Anthony and Ingo, adds > >paravirt_ops support for KVM guests enabling hypercall based pte updates, > >hypercall batching and cr3 caching. > > > > Could you post performance results for each optimization? I'm > particularly curious if the hypercall batching is very useful. Batched hypercall pte updates give 8.5% performance improvement on kernel compile: http://www.mail-archive.com/kvm-devel@lists.sourceforge.net/msg12395.html I can get separate results tomorrow or Monday, but I'm sure batching plays a significant role. For the kernel compile test, there is an average of 5 pte updates per batched hypercall. ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/