From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:37888) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dLxvb-00044X-5t for qemu-devel@nongnu.org; Fri, 16 Jun 2017 16:33:12 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dLxvW-0001vh-HM for qemu-devel@nongnu.org; Fri, 16 Jun 2017 16:33:11 -0400 Received: from mx1.redhat.com ([209.132.183.28]:49836) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dLxvW-0001tm-7y for qemu-devel@nongnu.org; Fri, 16 Jun 2017 16:33:06 -0400 Date: Fri, 16 Jun 2017 23:33:01 +0300 From: "Michael S. Tsirkin" Message-ID: <20170616232937-mutt-send-email-mst@kernel.org> References: <593FB550.6090903@intel.com> <26250da7-b394-4964-8842-5c45bbe85e09@redhat.com> <6547dfcf-ea3a-f5f6-222d-40ff274654df@redhat.com> <20170614180459-mutt-send-email-mst@kernel.org> <59422E91.7080407@intel.com> <20170616061949-mutt-send-email-mst@kernel.org> <40a12829-6e84-e63e-ac47-6f09cc85c3cc@redhat.com> <20170616181607-mutt-send-email-mst@kernel.org> <199a82f9-b122-08de-696a-04f7c6a369c6@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <199a82f9-b122-08de-696a-04f7c6a369c6@redhat.com> Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [virtio-dev] Re: [virtio-dev] Re: [virtio-dev] Re: [PATCH v1] virtio-net: enable configurable tx queue size List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Maxime Coquelin Cc: Jason Wang , Wei Wang , "virtio-dev@lists.oasis-open.org" , "stefanha@gmail.com" , "qemu-devel@nongnu.org" , "jan.scheurich@ericsson.com" , "armbru@redhat.com" , "marcandre.lureau@gmail.com" , "pbonzini@redhat.com" On Fri, Jun 16, 2017 at 07:04:27PM +0200, Maxime Coquelin wrote: >=20 >=20 > On 06/16/2017 05:19 PM, Michael S. Tsirkin wrote: > > On Fri, Jun 16, 2017 at 04:57:01PM +0800, Jason Wang wrote: > > >=20 > > >=20 > > > On 2017=E5=B9=B406=E6=9C=8816=E6=97=A5 11:22, Michael S. Tsirkin wr= ote: > > > > > I think the issues can be solved by VIRTIO_F_MAX_CHAIN_SIZE. > > > > >=20 > > > > > For now, how about splitting it into two series of patches: > > > > > 1) enable 1024 tx queue size for vhost-user, to let the users o= f vhost-user > > > > > to easily use 1024 queue size. > > > > Fine with me. 1) will get property from user but override it on > > > > !vhost-user. Do we need a protocol flag? It seems prudent but we = get > > > > back to cross-version migration issues that a04re still pending s= olution. > > > > Marc Andre, what's the status of that work? > > > >=20 > > > > > 2) enable VIRTIO_F_MAX_CHAIN_SIZE, to enhance robustness. > > > > Rather, to support it for more backends. > > >=20 > > > Ok, if we want to support different values of max chain size in the= future. > > > It would be problematic for migration of cross backends, consider t= he case > > > when migrating from 2048 (vhost-user) to 1024 (qemu/vhost-kernel). > > >=20 > > > Thanks > >=20 > > That's already a problem, and it's growing with each new feature. > > Maxime looked at supporting vhost-user backends cross-version migrati= on, > > I think we must merge some solution sooner rather than later, prefera= bly > > by the next release. > >=20 > > Maxime, any update here? Do we need a meeting to reach consensus? >=20 > No update, I haven't found time to progress on the topic yet. >=20 > For those who aren't aware of my initial proposal, you may find it here= : > https://www.spinics.net/linux/fedora/libvir/msg142668.html >=20 > If my understanding is correct, you were concerned about the complexity= of > my > proposal which involved too many layers. Your suggestion was to have a = tool > provided with qemu that would connect to vhost-user socket and query th= e > backend capabilities. > I'm not 100% clear how it would work, as the trend is to start the back= end > in > client mode, meaning QEMU creates the socket. In this case, should the = tool > create the socket and management tool request the backend to connect to= it? >=20 > I think it could make sense to have a meeting, but maybe we should firs= t > discuss the solutions on the list for efficiency. >=20 > For the delivery, what is QEMU v2.10 planned release date? >=20 > Note that my solution doesn't involve QEMU, so it would not be tight to= QEMU > release date. But, that doesn't mean it would be delivered sooner than > your solution. >=20 > Maxime I'd say let's go with your proposal (Solution 3 above). --=20 MST