From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kiszka Subject: Re: [PATCH v3 5/5] KVM: Allow host IRQ sharing for passed-through PCI 2.3 devices Date: Wed, 03 Nov 2010 18:27:50 +0100 Message-ID: <4CD19B96.2080603@siemens.com> References: <628f014fb1efb8e2208db03d13198ba301a3a34c.1288771873.git.jan.kiszka@web.de> <20101103084320.GF6772@redhat.com> <1288799867.3045.149.camel@x201> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: "Michael S. Tsirkin" , Avi Kivity , Marcelo Tosatti , kvm , Jan Kiszka To: Alex Williamson Return-path: Received: from thoth.sbs.de ([192.35.17.2]:20786 "EHLO thoth.sbs.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750801Ab0KCR2K (ORCPT ); Wed, 3 Nov 2010 13:28:10 -0400 In-Reply-To: <1288799867.3045.149.camel@x201> Sender: kvm-owner@vger.kernel.org List-ID: Am 03.11.2010 16:57, Alex Williamson wrote: > On Wed, 2010-11-03 at 10:43 +0200, Michael S. Tsirkin wrote: >> On Wed, Nov 03, 2010 at 09:11:16AM +0100, Jan Kiszka wrote: >>> From: Jan Kiszka >>> >>> PCI 2.3 allows to generically disable IRQ sources at device level. This >>> enables us to share IRQs of such devices between on the host side when >>> passing them to a guest. This feature is optional, user space has to >>> request it explicitly. >>> >>> Signed-off-by: Jan Kiszka >> >> >> I just realized something. >> With this patch, if guest ever looks at >> interrupt disable bit, it will go crazy as that bit goes on/off by >> itself. I guess we could have an ioctl to set/clear the bit on >> device, and have qemu call that on config write into command/status >> register. >> >> There's also something I don't completely unerstand with current code: >> how does interrupt sharing work? E.g. can assigned and emulated >> devices share an interrupt? > > I've been pondering this with VFIO too. There it seems to work, even > when I enable irqfd. The VFIO kernel/qemu driver needs to filter EOIs > based on whether the interrupt was actually asserted by the device, but > I think we're likely relying somewhat on interrupts being reasserted to > help us keep everything serviced. I don't think this filtering exists. The ack notifier that is fired on EOI matches the GSI, hitting anyone who is registered. I think the problem is that, while user space properly or's the input of all PCI devices on a IRQ line (e.g. in piix3_set_irq), kernel-side users apparently prefer to mess directly with the irqchip. Unless I'm missing something, that is long broken. Jan -- Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux