From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH] KVM: Use IRQF_ONESHOT for assigned device MSI interrupts Date: Mon, 11 Jun 2012 13:21:40 +0300 Message-ID: <20120611102139.GA31197@redhat.com> References: <1338570192.23475.25.camel@bling.home> <4FC8F867.7080103@siemens.com> <1338573558.23475.41.camel@bling.home> <4FC90961.8030701@siemens.com> <4FCB2359.9020505@redhat.com> <4FCC9EAC.9090007@siemens.com> <20120608143927.GA4949@redhat.com> <4FD21151.9090608@siemens.com> <4FD5C205.5070901@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Jan Kiszka , Thomas Gleixner , Alex Williamson , "kvm@vger.kernel.org" , "mtosatti@redhat.com" , "linux-kernel@vger.kernel.org" , "yongjie.ren@intel.com" To: Avi Kivity Return-path: Content-Disposition: inline In-Reply-To: <4FD5C205.5070901@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On Mon, Jun 11, 2012 at 01:01:41PM +0300, Avi Kivity wrote: > On 06/08/2012 05:50 PM, Jan Kiszka wrote: > > > >> > >> Pls correct me if I'm wrong. > > > > Well, IIRC, the "don't loop over all vcpus with IRQs or preemption > > disabled" was one argument against direct legacy interrupt injection as > > well. That's what I kept in mind from those discussions. Maybe Avi can > > comment on the current position. > > It's still my position. > > IMO we need something like struct gfn_to_hva_cache for interrupts. If > it's in the cache, we fast-path it from the interrupt handler. If not, > fall back to a workqueue and let it refill the cache. And you class the irqfd behaviour of injecting multicast with interrupts disabled a bug then? > -- > error compiling committee.c: too many arguments to function