From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:39624) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TGHgh-0006c2-NC for qemu-devel@nongnu.org; Mon, 24 Sep 2012 19:03:40 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1TGHgg-0005ax-Li for qemu-devel@nongnu.org; Mon, 24 Sep 2012 19:03:39 -0400 Received: from mx1.redhat.com ([209.132.183.28]:25936) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TGHgg-0005ai-Do for qemu-devel@nongnu.org; Mon, 24 Sep 2012 19:03:38 -0400 Date: Tue, 25 Sep 2012 01:04:53 +0200 From: "Michael S. Tsirkin" Message-ID: References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Subject: [Qemu-devel] [PATCH 08/14] virtio-net: switch tx to safe iov functions List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org, Jason Wang , Anthony Liguori , stefanha@linux.vnet.ibm.com, aurelien@aurel32.net Avoid mangling iovec manually: use safe iov_* functions. Signed-off-by: Michael S. Tsirkin --- hw/virtio-net.c | 31 +++++++++++++++++++------------ 1 file changed, 19 insertions(+), 12 deletions(-) diff --git a/hw/virtio-net.c b/hw/virtio-net.c index 6e6f5f3..23feb21 100644 --- a/hw/virtio-net.c +++ b/hw/virtio-net.c @@ -714,10 +714,10 @@ static int32_t virtio_net_flush_tx(VirtIONet *n, VirtQueue *vq) } while (virtqueue_pop(vq, &elem)) { - ssize_t ret, len = 0; + ssize_t ret, len; unsigned int out_num = elem.out_num; struct iovec *out_sg = &elem.out_sg[0]; - unsigned hdr_len; + struct iovec sg[VIRTQUEUE_MAX_SIZE]; /* hdr_len refers to the header received from the guest */ hdr_len = n->mergeable_rx_bufs ? @@ -729,18 +729,25 @@ static int32_t virtio_net_flush_tx(VirtIONet *n, VirtQueue *vq) exit(1); } - /* ignore the header if GSO is not supported */ - if (!n->has_vnet_hdr) { - out_num--; - out_sg++; - len += hdr_len; - } else if (n->mergeable_rx_bufs) { - /* tapfd expects a struct virtio_net_hdr */ - hdr_len -= sizeof(struct virtio_net_hdr); - out_sg->iov_len -= hdr_len; - len += hdr_len; + /* + * If host wants to see the guest header as is, we can + * pass it on unchanged. Otherwise, copy just the parts + * that host is interested in. + */ + assert(n->host_hdr_len <= n->guest_hdr_len); + if (n->host_hdr_len != n->guest_hdr_len) { + unsigned sg_num = iov_cpy(sg, ARRAY_SIZE(sg), + out_sg, out_num, + 0, n->host_hdr_len); + sg_num += iov_cpy(sg + sg_num, ARRAY_SIZE(sg) - sg_num, + out_sg, out_num, + n->guest_hdr_len, -1); + out_num = sg_num; + out_sg = sg; } + len = hdr_len; + ret = qemu_sendv_packet_async(&n->nic->nc, out_sg, out_num, virtio_net_tx_complete); if (ret == 0) { -- MST