From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:38923) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fXqRB-0001nQ-32 for qemu-devel@nongnu.org; Tue, 26 Jun 2018 12:03:26 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fXqR8-0006ex-Fw for qemu-devel@nongnu.org; Tue, 26 Jun 2018 12:03:25 -0400 Received: from mx3-rdu2.redhat.com ([66.187.233.73]:58662 helo=mx1.redhat.com) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1fXqR8-0006ei-CG for qemu-devel@nongnu.org; Tue, 26 Jun 2018 12:03:22 -0400 Date: Tue, 26 Jun 2018 18:03:16 +0200 From: Cornelia Huck Message-ID: <20180626180316.3723422f.cohuck@redhat.com> In-Reply-To: <20180626183706-mutt-send-email-mst@kernel.org> References: <20180623002628-mutt-send-email-mst@kernel.org> <20180623012406-mutt-send-email-mst@kernel.org> <20180626044650-mutt-send-email-mst@kernel.org> <20180626171732.5038f53f.cohuck@redhat.com> <20180626183706-mutt-send-email-mst@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [virtio-dev] Re: [PATCH] qemu: Introduce VIRTIO_NET_F_STANDBY feature bit to virtio_net List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Michael S. Tsirkin" Cc: "Samudrala, Sridhar" , Siwei Liu , Alexander Duyck , virtio-dev@lists.oasis-open.org, aaron.f.brown@intel.com, Jiri Pirko , Jakub Kicinski , Netdev , qemu-devel@nongnu.org, virtualization@lists.linux-foundation.org, konrad.wilk@oracle.com, boris.ostrovsky@oracle.com, Joao Martins , Venu Busireddy , vijay.balakrishna@oracle.com On Tue, 26 Jun 2018 18:38:51 +0300 "Michael S. Tsirkin" wrote: > On Tue, Jun 26, 2018 at 05:17:32PM +0200, Cornelia Huck wrote: > > On Tue, 26 Jun 2018 04:50:25 +0300 > > "Michael S. Tsirkin" wrote: > > =20 > > > On Mon, Jun 25, 2018 at 10:54:09AM -0700, Samudrala, Sridhar wrote: = =20 > > > > > > > > Might not neccessarily be something wrong, but it's very li= mited to > > > > > > > > prohibit the MAC of VF from changing when enslaved by failo= ver. =20 > > > > > > > You mean guest changing MAC? I'm not sure why we prohibit tha= t. =20 > > > > > > I think Sridhar and Jiri might be better person to answer it. My > > > > > > impression was that sync'ing the MAC address change between all= 3 > > > > > > devices is challenging, as the failover driver uses MAC address= to > > > > > > match net_device internally. =20 > > > >=20 > > > > Yes. The MAC address is assigned by the hypervisor and it needs to = manage the movement > > > > of the MAC between the PF and VF.=C2=A0 Allowing the guest to chang= e the MAC will require > > > > synchronization between the hypervisor and the PF/VF drivers. Most = of the VF drivers > > > > don't allow changing guest MAC unless it is a trusted VF. =20 > > >=20 > > > OK but it's a policy thing. Maybe it's a trusted VF. Who knows? > > > For example I can see host just > > > failing VIRTIO_NET_CTRL_MAC_ADDR_SET if it wants to block it. > > > I'm not sure why VIRTIO_NET_F_STANDBY has to block it in the guest. > > > =20 > >=20 > > So, what I get from this is that QEMU needs to be able to control all > > of standby, uuid, and mac to accommodate the different setups > > (respectively have libvirt/management software set it up). Is the host > > able to find out respectively define whether a VF is trusted? =20 >=20 > You do it with ip link I think but QEMU doesn't normally do this, > it relies on libvirt to poke at host kernel and supply the info. >=20 Ok, that makes me conclude that we definitely need to involve the libvirt folks before we proceed further with defining QEMU interfaces.