From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [RFC] Paravirt timer for KVM Date: Tue, 16 Oct 2007 09:31:58 +0200 Message-ID: <471468EE.1030706@qumranet.com> References: <5d6222a80710120908s6b1f5845head84e7b7a463cd1@mail.gmail.com> <470FB8AB.9030101@codemonkey.ws> <5d6222a80710150958y31338c2ag3a391390b13788da@mail.gmail.com> <47139E6F.7030704@qumranet.com> <5d6222a80710151109m5376449foc6be5b687c469a2b@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: kvm-devel , Jeremy Fitzhardinge To: Glauber de Oliveira Costa Return-path: In-Reply-To: <5d6222a80710151109m5376449foc6be5b687c469a2b-JsoAwUIsXosN+BqQ9rBEUg@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 Glauber de Oliveira Costa wrote: > On 10/15/07, Avi Kivity wrote: > >> If we want to make the clocksource useful for Windows we need to go >> through the apic as that will allow the TPR to mask the interrupt when >> Windows isn't ready to receive it. However I don't know whether it is >> possible to integrate a paravirt clocksource into Windows. >> > > My politically-correct side told me to say that the solution that > works also in Windows is the best one ;-) > > In this case, I'm not at all sure Windows can benefit from a paravirt timesource, so it might be better to concentrate on Linux. I'll ask the Windows people here. >> For Linux, we might as well go further and provide a completely >> paravirtualized irqchip instead of trying to integrate a paravirtualized >> clocksource into a fullyvirtualized irqchip. >> >> > > Yes, definitely. In your opinion, what's the size of the gap between > the "we can", and the "we should" here? > At present the APIC requires an exit for each level-triggered interrupt. We could reduce the number of exits to zero given sufficient cleverness. This would be beneficial to non-streaming I/O workloads. So yes, I think it's worthwhile. -- Do not meddle in the internals of kernels, for they are subtle and quick to panic. ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/