From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:34476) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1edTCC-0006I5-7I for qemu-devel@nongnu.org; Sun, 21 Jan 2018 22:54:57 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1edTC9-0005Sd-6N for qemu-devel@nongnu.org; Sun, 21 Jan 2018 22:54:56 -0500 Received: from mx1.redhat.com ([209.132.183.28]:53296) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1edTC8-0005Rg-WF for qemu-devel@nongnu.org; Sun, 21 Jan 2018 22:54:53 -0500 References: <20180110161438.GA28096@stefanha-x1.localdomain> <20180111152345.GA7353@stefanha-x1.localdomain> <86106573-422b-fe4c-ec15-dad0edf05880@redhat.com> <20180112101807.GE7356@stefanha-x1.localdomain> <20180115135620.GG13238@stefanha-x1.localdomain> <9fad276a-d17b-6a45-6cd6-50899934b7a1@redhat.com> <20180118105103.GC19831@stefanha-x1.localdomain> <78448708-8d53-70ca-d079-424caabc3a35@redhat.com> <20180119172049.GA29973@stefanha-x1.localdomain> From: Jason Wang Message-ID: <9c05c203-7d1f-dbea-5577-4d85c42e82ee@redhat.com> Date: Mon, 22 Jan 2018 11:54:41 +0800 MIME-Version: 1.0 In-Reply-To: <20180119172049.GA29973@stefanha-x1.localdomain> Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] vhost-pci and virtio-vhost-user List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Stefan Hajnoczi Cc: wei.w.wang@intel.com, qemu-devel@nongnu.org, Stefan Hajnoczi On 2018=E5=B9=B401=E6=9C=8820=E6=97=A5 01:20, Stefan Hajnoczi wrote: >> I don't propose any new idea. I just want to know what's the advantage= of >> vhost-pci over zerocopy. Both needs one time of copy, the difference i= s the >> vhost-pci did it inside a guest but zerocopy did in on host. > Exitless VM2VM communication is desirable if you cannot run software on > the host or if both endpoints are already in VMs. In that case running > one thing in a VM and another on the host doesn't make sense. Well, I must have missed anything, I don't see why we can not run=20 virtio-net backend on host. Especially it only does L2 stuffs, higher=20 level of service could be provided by another VM for sure. So it looks=20 to me virtio-vhost-user is just a split device implementation which is=20 irreverent to the service it could provide. Maybe you can provide a concrete examples of virtio-vhost-user and its=20 advantages? > > The obvious environment where this applies is in the cloud where > everything is a VM. So a typical setup makes the VMs can already talk to each other through=20 ethernet(virtio-net). Virtio-vhost-user looks much less flexible than=20 exist stuffs. The only possible advantage of virtio-vhost-user is its=20 performance or security which still need to be proved. Thanks