From patchwork Mon Sep 24 23:05:12 2012 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Michael S. Tsirkin" X-Patchwork-Id: 186607 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTPS id 0A3392C007C for ; Tue, 25 Sep 2012 09:50:37 +1000 (EST) Received: from localhost ([::1]:48213 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TGIQ7-0008Ma-9C for incoming@patchwork.ozlabs.org; Mon, 24 Sep 2012 19:50:35 -0400 Received: from eggs.gnu.org ([208.118.235.92]:39809) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TGHh7-0007uR-95 for qemu-devel@nongnu.org; Mon, 24 Sep 2012 19:04:06 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1TGHh5-0005fs-N5 for qemu-devel@nongnu.org; Mon, 24 Sep 2012 19:04:05 -0400 Received: from mx1.redhat.com ([209.132.183.28]:10386) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TGHh5-0005fk-9t for qemu-devel@nongnu.org; Mon, 24 Sep 2012 19:04:03 -0400 Received: from int-mx11.intmail.prod.int.phx2.redhat.com (int-mx11.intmail.prod.int.phx2.redhat.com [10.5.11.24]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id q8ON3d7P032680 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 24 Sep 2012 19:03:39 -0400 Received: from redhat.com (vpn1-7-132.ams2.redhat.com [10.36.7.132]) by int-mx11.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id q8ON3aNd001669; Mon, 24 Sep 2012 19:03:37 -0400 Date: Tue, 25 Sep 2012 01:05:12 +0200 From: "Michael S. Tsirkin" To: qemu-devel@nongnu.org, Jason Wang , Anthony Liguori , stefanha@linux.vnet.ibm.com, aurelien@aurel32.net Message-ID: References: MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: X-Mutt-Fcc: =sent X-Scanned-By: MIMEDefang 2.68 on 10.5.11.24 X-detected-operating-system: by eggs.gnu.org: Genre and OS details not recognized. X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PATCH 14/14] virtio-net: enable mrg buf header in tap on linux X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Sender: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Modern linux supports arbitrary header size, which makes it possible to pass mrg buf header to tap directly without iovec mangling. Use this capability when it is there. This removes the need to deal with it in vhost-net as we do now. Signed-off-by: Michael S. Tsirkin --- hw/vhost_net.c | 13 ------------- hw/virtio-net.c | 26 ++++++++++++++++++-------- 2 files changed, 18 insertions(+), 21 deletions(-) diff --git a/hw/vhost_net.c b/hw/vhost_net.c index df2c4a3..8241601 100644 --- a/hw/vhost_net.c +++ b/hw/vhost_net.c @@ -150,10 +150,6 @@ int vhost_net_start(struct vhost_net *net, if (r < 0) { goto fail_notifiers; } - if (net->dev.acked_features & (1 << VIRTIO_NET_F_MRG_RXBUF)) { - tap_set_vnet_hdr_len(net->nc, - sizeof(struct virtio_net_hdr_mrg_rxbuf)); - } r = vhost_dev_start(&net->dev, dev); if (r < 0) { @@ -179,9 +175,6 @@ fail: } net->nc->info->poll(net->nc, true); vhost_dev_stop(&net->dev, dev); - if (net->dev.acked_features & (1 << VIRTIO_NET_F_MRG_RXBUF)) { - tap_set_vnet_hdr_len(net->nc, sizeof(struct virtio_net_hdr)); - } fail_start: vhost_dev_disable_notifiers(&net->dev, dev); fail_notifiers: @@ -199,18 +192,12 @@ void vhost_net_stop(struct vhost_net *net, } net->nc->info->poll(net->nc, true); vhost_dev_stop(&net->dev, dev); - if (net->dev.acked_features & (1 << VIRTIO_NET_F_MRG_RXBUF)) { - tap_set_vnet_hdr_len(net->nc, sizeof(struct virtio_net_hdr)); - } vhost_dev_disable_notifiers(&net->dev, dev); } void vhost_net_cleanup(struct vhost_net *net) { vhost_dev_cleanup(&net->dev); - if (net->dev.acked_features & (1 << VIRTIO_NET_F_MRG_RXBUF)) { - tap_set_vnet_hdr_len(net->nc, sizeof(struct virtio_net_hdr)); - } g_free(net); } #else diff --git a/hw/virtio-net.c b/hw/virtio-net.c index a7c8bb3..7558eab 100644 --- a/hw/virtio-net.c +++ b/hw/virtio-net.c @@ -228,6 +228,20 @@ static int peer_has_ufo(VirtIONet *n) return n->has_ufo; } +static void virtio_net_set_mrg_rx_bufs(VirtIONet *n, int mergeable_rx_bufs) +{ + n->mergeable_rx_bufs = mergeable_rx_bufs; + + n->guest_hdr_len = n->mergeable_rx_bufs ? + sizeof(struct virtio_net_hdr_mrg_rxbuf) : sizeof(struct virtio_net_hdr); + + if (peer_has_vnet_hdr(n) && + tap_has_vnet_hdr_len(n->nic->nc.peer, n->guest_hdr_len)) { + tap_set_vnet_hdr_len(n->nic->nc.peer, n->guest_hdr_len); + n->host_hdr_len = n->guest_hdr_len; + } +} + static uint32_t virtio_net_get_features(VirtIODevice *vdev, uint32_t features) { VirtIONet *n = to_virtio_net(vdev); @@ -280,9 +294,7 @@ static void virtio_net_set_features(VirtIODevice *vdev, uint32_t features) { VirtIONet *n = to_virtio_net(vdev); - n->mergeable_rx_bufs = !!(features & (1 << VIRTIO_NET_F_MRG_RXBUF)); - n->guest_hdr_len = n->mergeable_rx_bufs ? - sizeof(struct virtio_net_hdr_mrg_rxbuf) : sizeof(struct virtio_net_hdr); + virtio_net_set_mrg_rx_bufs(n, !!(features & (1 << VIRTIO_NET_F_MRG_RXBUF))); if (n->has_vnet_hdr) { tap_set_offload(n->nic->nc.peer, @@ -897,9 +909,8 @@ static int virtio_net_load(QEMUFile *f, void *opaque, int version_id) qemu_get_buffer(f, n->mac, ETH_ALEN); n->tx_waiting = qemu_get_be32(f); - n->mergeable_rx_bufs = qemu_get_be32(f); - n->guest_hdr_len = n->mergeable_rx_bufs ? - sizeof(struct virtio_net_hdr_mrg_rxbuf) : sizeof(struct virtio_net_hdr); + + virtio_net_set_mrg_rx_bufs(n, qemu_get_be32(f)); if (version_id >= 3) n->status = qemu_get_be16(f); @@ -1042,8 +1053,7 @@ VirtIODevice *virtio_net_init(DeviceState *dev, NICConf *conf, n->tx_waiting = 0; n->tx_burst = net->txburst; - n->mergeable_rx_bufs = 0; - n->guest_hdr_len = sizeof(struct virtio_net_hdr); + virtio_net_set_mrg_rx_bufs(n, 0); n->promisc = 1; /* for compatibility */ n->mac_table.macs = g_malloc0(MAC_TABLE_ENTRIES * ETH_ALEN);