From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alex Williamson Subject: Re: PCI-Passthrough: VF ends up with no IRQ assigned on the physical machine Date: Mon, 26 Mar 2012 16:38:03 -0600 Message-ID: <1332801483.29722.70.camel@bling.home> References: Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: kvm@vger.kernel.org To: Alexander Lyakas Return-path: Received: from mx1.redhat.com ([209.132.183.28]:15927 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751659Ab2CZWiG (ORCPT ); Mon, 26 Mar 2012 18:38:06 -0400 In-Reply-To: Sender: kvm-owner@vger.kernel.org List-ID: On Mon, 2012-03-26 at 18:34 +0200, Alexander Lyakas wrote: > Greetings everybody, > I am running stock ubuntu-natty 2.6.38-8, with KVM 0.14.0 and libvirt Any chance you can try upstream qemu-kvm? Interrupts, including msi-x have been given an overhaul recently. Thanks, Alex > 0.8.8. I am using Intel 82599EB dual-port 10Gb NIC with VFs spawned by > each one of the PFs. Those VFs are further assigned to KVM instances > with PCI-passthrough using libvirt. Each instance receives 4 VFs (two > VFs from each PF). > Usually when KVM starts, messages like these appear in the kernel log > of the physical machine: > syslog.ubuntu-sata-51 146907 Mar 25 18:19:51 ubuntu-sata-51 kernel: > [51267.029108] pci-stub 0000:03:13.1: claimed by stub > syslog.ubuntu-sata-51 146908 Mar 25 18:19:51 ubuntu-sata-51 kernel: ...