From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ingo Molnar Subject: Re: [announce] [patch] KVM paravirtualization for Linux Date: Mon, 8 Jan 2007 10:43:59 +0100 Message-ID: <20070108094359.GB30482@elte.hu> References: <20070105215223.GA5361@elte.hu> <45A0E586.50806@qumranet.com> <20070107174416.GA14607@elte.hu> <45A1FF4E.1020106@qumranet.com> <20070108083935.GB18259@elte.hu> <45A20A0A.70403@qumranet.com> <20070108091819.GB26587@elte.hu> <45A20F60.20207@qumranet.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: kvm-devel , linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org Return-path: To: Avi Kivity Content-Disposition: inline In-Reply-To: <45A20F60.20207-atKUWr5tajBWk0Htik3J/w@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 * Avi Kivity wrote: > Looks like a lot of complexity for very little gain. I'm not sure > what the vmwrite cost is, cut it can't be that high compared to > vmexit. while i disagree with characterising one extra parameter passed down plus one extra branch as 'a lot of complexity', i agree that making the flush a NOP on VMX is even better. (if it's possible without breaking future hardware) I guess you are right that the only true cost here is the vmwrite cost, and that there's no impact on CR3 flushing (because it happens unconditionally). Forget about this patch. Ingo ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV