From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:35798) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eFYKy-0005Jk-3P for qemu-devel@nongnu.org; Thu, 16 Nov 2017 23:33:09 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eFYKu-0001JI-Vp for qemu-devel@nongnu.org; Thu, 16 Nov 2017 23:33:08 -0500 Received: from mx1.redhat.com ([209.132.183.28]:56892) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1eFYKu-0001Gp-Lk for qemu-devel@nongnu.org; Thu, 16 Nov 2017 23:33:04 -0500 Date: Fri, 17 Nov 2017 06:32:58 +0200 From: "Michael S. Tsirkin" Message-ID: <20171117063024-mutt-send-email-mst@kernel.org> References: <5A0D27D8.6050506@huawei.com> <2e0ddd76-3903-d715-9a86-73f1286322e7@redhat.com> <33183CC9F5247A488A2544077AF19020DA46E2A6@DGGEMA505-MBS.china.huawei.com> <6aaaca2b-81ef-a1c1-4745-830d008b3770@redhat.com> <5A0D5B1B.3080708@huawei.com> <55f2edb1-1fe3-d03f-7f4f-418a3b15997a@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <55f2edb1-1fe3-d03f-7f4f-418a3b15997a@redhat.com> Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [Question] why need to start all queues in vhost_net_start List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Jason Wang Cc: "Longpeng (Mike)" , "Huangweidong (C)" , "stefanha@redhat.com" , "qemu-devel@nongnu.org" , "Wangjing (King, Euler)" , "Gonglei (Arei)" , "Longpeng(Mike)" On Thu, Nov 16, 2017 at 08:04:34PM +0800, Jason Wang wrote: >=20 >=20 > On 2017=E5=B9=B411=E6=9C=8816=E6=97=A5 17:32, Longpeng (Mike) wrote: > > Hi Jason, > >=20 > > On 2017/11/16 17:13, Jason Wang wrote: > >=20 > > >=20 > > > On 2017=E5=B9=B411=E6=9C=8816=E6=97=A5 17:01, Gonglei (Arei) wrote: > > > > No, Windows guest + vhost-user/DPDK. > > > >=20 > > > > BTW pls see virtio spec in : > > > >=20 > > > > "If VIRTIO_NET_F_MQ is negotiated, each of receiveq1. . .receiveq= N that will > > > > be used SHOULD be populated > > > > with receive buffers." > > > >=20 > > > > It is not mandatory that all queues must be initialized. > > > I think not, since it said we should fill receive buffers for each = queue which > > > means we should initialize all queues. May Michael can clarify on t= his. > > >=20 > >=20 > > I think this doesn't matter, but QEMU should consider this scenario..= . > >=20 > > For example, if one queues isn't initialized (Windows guest), the vri= ng.avail=3D0, > > so vq->desc_phys=3D0, then vq->desc=3D'a avail HVA'(which is the star= t addr of pc.ram). > >=20 > > vq->desc_size =3D s =3D l =3D virtio_queue_get_desc_size(vdev, i= dx); > > vq->desc_phys =3D a =3D virtio_queue_get_desc_addr(vdev, idx); > > vq->desc =3D vhost_memory_map(dev, a, &l, 0); > > if (!vq->desc || l !=3D s) { > > r =3D -ENOMEM; > > goto fail_alloc_desc; > > } > > ..... > > r =3D vhost_virtqueue_set_addr(dev, vq, vhost_vq_index, dev->log= _enabled); > > if (r < 0) { > > r =3D -errno; > > goto fail_alloc; > > } > >=20 > > Then the HVA is send to the vhost-user. > >=20 > > I think this is wrong, because the '0' here means guest driver doesn'= t init this > > queues, it should not be used to calculate the HVA for this vq. >=20 > Yes, workaround is not hard if windows driver won't use the left 3 queu= es > any more. But we should have a complete solution. The main problem is w= hen > vhost need to be started. For legacy device, there's no easy way to det= ect > whether or not a specific virtqueue is ready to be used. For modern dev= ice, > we can probably do this through queue_enable (but this is not implement= ed in > current code). >=20 > Thanks What isn't implemented? Spec is quite explicit: Client must only process each ring when it is started. Client must only pass data between the ring and the backend, when the ring is enabled. and later: Client must start ring upon receiving a kick (that is, detecting that fil= e descriptor is readable) on the descriptor specified by VHOST_USER_SET_VRING_KICK, and stop ring upon receiving VHOST_USER_GET_VRING_BASE. Does someone kick unused rings? What entity does this? > >=20 > > > Thanks > > >=20 > > > > Thanks, > > > > -Gonglei > > > >=20 > > >=20 > > >=20 > > >=20 > > > . > > >=20 > >=20