Re: vhost-net patches

["Michael S. Tsirkin" <mst@redhat.com>]

On Fri, Oct 23, 2009 at 09:23:40AM -0700, Shirley Ma wrote:
> Hello Michael,
> 
> Some initial vhost test netperf results on my T61 laptop from the
> working tap device are here, latency has been significant decreased, but
> throughput from guest to host has huge regression. I also hit guest
> skb_xmit panic.
> 
> netperf TCP_STREAM, default setup, 60 secs run
> 
> guest->host drops from 3XXXMb/s to 1XXXMb/s (regression)
> host->guest increases from 3XXXMb/s to 4XXXXMb/s 
> 
> TCP_RR, 60 secs run (very impressive)
> 
> guest->host trans/s increases from 2XXX/s to 13XXX/s
> host->guest trans/s increases from 2XXX/s to 13XXX/s
> 
> Thanks
> Shirley

Shirley, could you please test the following patch?
It is surprising to me that it should improve
performance, but seems to do this in my setup.
Please comment.


diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 30708c6..67bfc08 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -775,7 +775,7 @@ void vhost_no_notify(struct vhost_virtqueue *vq)
 
 int vhost_init(void)
 {
-	vhost_workqueue = create_workqueue("vhost");
+	vhost_workqueue = create_singlethread_workqueue("vhost");
 	if (!vhost_workqueue)
 		return -ENOMEM;
 	return 0;
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index a140dad..49026bb 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -106,10 +106,14 @@ static void handle_tx(struct vhost_net *net)
 		.msg_flags = MSG_DONTWAIT,
 	};
 	size_t len, total_len = 0;
-	int err;
+	int err, wmem;
 	size_t hdr_size;
 	struct socket *sock = rcu_dereference(vq->private_data);
-	if (!sock || !sock_writeable(sock->sk))
+	if (!sock)
+		return;
+
+	wmem = atomic_read(&sock->sk->sk_wmem_alloc);
+	if (wmem >= sock->sk->sk_sndbuf)
 		return;
 
 	use_mm(net->dev.mm);