From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?ISO-8859-15?Q?Andreas_F=E4rber?= Subject: Re: [PATCH v3 4/4] kvm: i386: Add classic PCI device assignment Date: Thu, 06 Sep 2012 18:06:42 +0200 Message-ID: <5048CA12.8060200@suse.de> References: <825e653c9cfe9d8e26185917cbe1f1dd7ae299e2.1346048917.git.jan.kiszka@web.de> <503CA542.7030707@siemens.com> <503FB128.4070709@siemens.com> <5048625F.2060706@siemens.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Avi Kivity , Marcelo Tosatti , kvm@vger.kernel.org, "Michael S. Tsirkin" , qemu-devel@nongnu.org, Blue Swirl , Alex Williamson , Peter Maydell To: Jan Kiszka Return-path: Received: from cantor2.suse.de ([195.135.220.15]:54990 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932082Ab2IFQGr (ORCPT ); Thu, 6 Sep 2012 12:06:47 -0400 In-Reply-To: <5048625F.2060706@siemens.com> Sender: kvm-owner@vger.kernel.org List-ID: Am 06.09.2012 10:44, schrieb Jan Kiszka: > On 2012-08-30 20:30, Jan Kiszka wrote: >> This adds PCI device assignment for i386 targets using the classic K= VM >> interfaces. This version is 100% identical to what is being maintain= ed >> in qemu-kvm for several years and is supported by libvirt as well. I= t is >> expected to remain relevant for another couple of years until kernel= s >> without full-features and performance-wise equivalent VFIO support a= re >> obsolete. >> >> A refactoring to-do that should be done in-tree is to model MSI and >> MSI-X support via the generic PCI layer, similar to what VFIO is alr= eady >> doing for MSI-X. This should improve the correctness and clean up th= e >> code from duplicate logic. >> >> Signed-off-by: Jan Kiszka >> --- >> >> Changes in v3: >> - addressed comment by Peter (changed device name to kvm-pci-assign= + >> alias) >> - addressed (most) comments by Michael >> - fixed INT pin regression >=20 > Does someone _disagree_ that there are no open (and reasonably solvab= le) > issues and that this can now be merged through uq/master? My implicit suggestion was to add a notice that new patch contributions to the file from date yyyy-mm-dd on would be declared GPLv2+, as Paolo has done elsewhere. That would limit the amount of people to ask for a potential relicensing attempt. =46or the record, Anthony explained on IRC that the code originated fro= m Xen originally and thus qemu-kvm.git does not contain the full history anyway and that pulling in the Mercurial file history and replaying the KVM history on top was too difficult, therefore this patch with a singl= e SoB by Jan. Andreas --=20 SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 N=FCrnberg, Germany GF: Jeff Hawn, Jennifer Guild, Felix Imend=F6rffer; HRB 16746 AG N=FCrn= berg From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:49135) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1T9ebW-0001uE-DY for qemu-devel@nongnu.org; Thu, 06 Sep 2012 12:06:55 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1T9ebQ-0003u6-2F for qemu-devel@nongnu.org; Thu, 06 Sep 2012 12:06:54 -0400 Received: from cantor2.suse.de ([195.135.220.15]:54987 helo=mx2.suse.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1T9ebP-0003tp-PI for qemu-devel@nongnu.org; Thu, 06 Sep 2012 12:06:47 -0400 Message-ID: <5048CA12.8060200@suse.de> Date: Thu, 06 Sep 2012 18:06:42 +0200 From: =?ISO-8859-15?Q?Andreas_F=E4rber?= MIME-Version: 1.0 References: <825e653c9cfe9d8e26185917cbe1f1dd7ae299e2.1346048917.git.jan.kiszka@web.de> <503CA542.7030707@siemens.com> <503FB128.4070709@siemens.com> <5048625F.2060706@siemens.com> In-Reply-To: <5048625F.2060706@siemens.com> Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v3 4/4] kvm: i386: Add classic PCI device assignment List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Jan Kiszka Cc: Peter Maydell , kvm@vger.kernel.org, "Michael S. Tsirkin" , Marcelo Tosatti , qemu-devel@nongnu.org, Blue Swirl , Alex Williamson , Avi Kivity Am 06.09.2012 10:44, schrieb Jan Kiszka: > On 2012-08-30 20:30, Jan Kiszka wrote: >> This adds PCI device assignment for i386 targets using the classic KVM >> interfaces. This version is 100% identical to what is being maintained >> in qemu-kvm for several years and is supported by libvirt as well. It = is >> expected to remain relevant for another couple of years until kernels >> without full-features and performance-wise equivalent VFIO support are >> obsolete. >> >> A refactoring to-do that should be done in-tree is to model MSI and >> MSI-X support via the generic PCI layer, similar to what VFIO is alrea= dy >> doing for MSI-X. This should improve the correctness and clean up the >> code from duplicate logic. >> >> Signed-off-by: Jan Kiszka >> --- >> >> Changes in v3: >> - addressed comment by Peter (changed device name to kvm-pci-assign + >> alias) >> - addressed (most) comments by Michael >> - fixed INT pin regression >=20 > Does someone _disagree_ that there are no open (and reasonably solvable= ) > issues and that this can now be merged through uq/master? My implicit suggestion was to add a notice that new patch contributions to the file from date yyyy-mm-dd on would be declared GPLv2+, as Paolo has done elsewhere. That would limit the amount of people to ask for a potential relicensing attempt. For the record, Anthony explained on IRC that the code originated from Xen originally and thus qemu-kvm.git does not contain the full history anyway and that pulling in the Mercurial file history and replaying the KVM history on top was too difficult, therefore this patch with a single SoB by Jan. Andreas --=20 SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 N=FCrnberg, Germany GF: Jeff Hawn, Jennifer Guild, Felix Imend=F6rffer; HRB 16746 AG N=FCrnbe= rg