From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH] Remove redundant MSR save/restore Date: Wed, 25 Apr 2007 12:28:39 +0300 Message-ID: <462F1F47.8020209@qumranet.com> References: <10EA09EFD8728347A513008B6B0DA77A01599E01@pdsmsx411.ccr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: kvm-devel To: "Dong, Eddie" Return-path: In-Reply-To: <10EA09EFD8728347A513008B6B0DA77A01599E01-wq7ZOvIWXbNpB2pF5aRoyrfspsVTdybXVpNB7YpNyf8@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Errors-To: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Id: kvm.vger.kernel.org Dong, Eddie wrote: > Avi: > Thanks! > Another found is that HOST_FS/GS_BASE & MSR_FS/GS_BASE > save/restore is quit expansive. I am not sure why it needs to be > save/restored given that the VMCS regster will be loaded into FS/gs at > VM EXIT time automatically by HW and will not be changed during guest > execution time. The MSR is also same (guest change will be trap and > emulated in guest context only). > If I remove HOST_FS/GS_BASE save/restore out to be only matter > to userspace, I can get 15% performance gain for KB. But I am not sure > yet if there is any potential issue. > > thanks, eddie > I don't see issues with moving them outside the loop, as long as preemption is disabled. -- error compiling committee.c: too many arguments to function ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/