From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:51770) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XQznF-0002Gb-R4 for qemu-devel@nongnu.org; Mon, 08 Sep 2014 10:19:51 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XQznB-0002HJ-9c for qemu-devel@nongnu.org; Mon, 08 Sep 2014 10:19:45 -0400 Received: from mx1.redhat.com ([209.132.183.28]:50542) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XQznB-0002H7-1m for qemu-devel@nongnu.org; Mon, 08 Sep 2014 10:19:41 -0400 Message-ID: <1410185972.2982.136.camel@ul30vt.home> From: Alex Williamson Date: Mon, 08 Sep 2014 08:19:32 -0600 In-Reply-To: <540D74F9.50903@redhat.com> References: <20140904105223.336503578@de.ibm.com> <54097287.1000006@suse.de> <20140905113954.GB17325@tuxmaker.boeblingen.de.ibm.com> <540A44F2.3040803@suse.de> <540D74F9.50903@redhat.com> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [RFC][patch 0/6] pci pass-through support for qemu/KVM on s390 List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini Cc: linux-s390@vger.kernel.org, frank.blaschka@de.ibm.com, kvm@vger.kernel.org, aik@ozlabs.ru, Frank Blaschka , Alexander Graf , qemu-devel@nongnu.org On Mon, 2014-09-08 at 11:20 +0200, Paolo Bonzini wrote: > Il 06/09/2014 01:19, Alexander Graf ha scritto: > >> > 1) interpretive execution of pci load/store instruction. If we use this function > >> > pci access does not get intercepted (no SIE exit) but is handled via microcode. > >> > To enable this we have to disable zpci device and enable it again with information > >> > from the SIE control block. > > Hrm. So how about you create a special vm ioctl for KVM that allows you > > to attach a VFIO device fd into the KVM VM context? Then the default > > would stay "accessible by mmap traps", but we could accelerate it with KVM. > > There is already KVM_DEV_VFIO_GROUP_ADD and KVM_DEV_VFIO_GROUP_DEL. > > Right now, they result in a call to kvm_arch_register_noncoherent_dma or > kvm_arch_unregister_noncoherent_dma, but you can add more hooks. Eric Auger is also working on a patch series to do IRQ forward control on ARM via the kvm-vfio pseudo device, extending the interface to register VFIO device fds. Sounds like that may be a good path to follow here too. Thanks, Alex