From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Gregory Haskins" Subject: Re: Fwd: Re: [PATCH] Add irqdevice interface + userint implementation Date: Tue, 10 Apr 2007 13:58:01 -0400 Message-ID: <461B97DA.BA47.005A.0@novell.com> References: <4614FF5C.BA47.005A.0@novell.com> <46189005.2090609@qumranet.com> <461A415C.BA47.005A.0@novell.com> <461B4241.5030101@qumranet.com> <461B3E61.BA47.005A.0@novell.com> <461B7C81.7040102@qumranet.com> <461B64E4.BA47.005A.0@novell.com> <461BA363.5080308@qumranet.com> <461B7755.BA47.005A.0@novell.com> <461B7AC6.BA47.005A.0@novell.com> <461BB738.7040206@qumranet.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org To: "Avi Kivity" Return-path: In-Reply-To: <461BB738.7040206-atKUWr5tajBWk0Htik3J/w@public.gmane.org> Content-Disposition: inline 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 >>> On Tue, Apr 10, 2007 at 12:11 PM, in message <461BB738.7040206-atKUWr5tajBWk0Htik3J/w@public.gmane.org>, Avi Kivity wrote: > > > It's more complex than that; Isn't it always ;) > you need to program the hardware to exit on > a specific threshold level of tpr values (e.g. exit when tpr <= n). > Agreed, but don't forget the TPR/CR8 will always be an exposed interface of the LAPIC and therefore readily available for factoring into the algorithm. I just am not there yet. > It's a little sad to encode a complex condition (wait for nmi window > opening/wait for irq window opening/wait for irq window opening with tpr > <= n) in irqdevice, then decode it in the vcpu and program the hardware > accordingly. Agreed, but keep in mind that whatever we make the irqdevice do, it will have to do it in both the QEMU and in-kernel models if we want true dynamic ABI support for old userspace. I was trying to reduce the duplication of code in each model, but perhaps it's no more complex either way...TBD. Regards, -Greg ------------------------------------------------------------------------- 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