From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCHv3] KVM: optimize apic interrupt delivery Date: Sun, 16 Sep 2012 14:04:56 +0300 Message-ID: <5055B258.7070902@redhat.com> References: <20120913090059.GZ20907@redhat.com> <5051B598.4000708@siemens.com> <20120913103311.GA2924@redhat.com> <5051B72E.8040105@siemens.com> <20120916105655.GA25288@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Jan Kiszka , Gleb Natapov , "kvm@vger.kernel.org" , "mtosatti@redhat.com" To: "Michael S. Tsirkin" Return-path: Received: from mx1.redhat.com ([209.132.183.28]:18111 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751022Ab2IPLFB (ORCPT ); Sun, 16 Sep 2012 07:05:01 -0400 In-Reply-To: <20120916105655.GA25288@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On 09/16/2012 01:56 PM, Michael S. Tsirkin wrote: > On Thu, Sep 13, 2012 at 12:36:30PM +0200, Jan Kiszka wrote: >> On 2012-09-13 12:33, Gleb Natapov wrote: >> >> >> >> So, this can be the foundation for direct MSI delivery as well, right? >> >> >> > What do you mean by "direct MSI delivery"? kvm_irq_delivery_to_apic() is >> > called by MSI. If you mean delivery from irq context, then yes, mst >> > plans to do so. >> >> Yes, that's what I was aiming at. >> >> Jan > > Avi, I've been debating with self whether it's still worth it > to bounce interrupts that don't get optimized such as > broadcasts, out to thread? Maybe we can just do direct delivery > unconditionally? Guests don't seem to use broadcasts at all widely ... Malicious guests do, all the time. -- error compiling committee.c: too many arguments to function