vhost_net: use vhost_add_used_and_signal_n() in vhost_zerocopy_signal_used()

We tend to batch the used adding and signaling in vhost_zerocopy_callback()
which may result more than 100 used buffers to be updated in
vhost_zerocopy_signal_used() in some cases. So switch to use
vhost_add_used_and_signal_n() to avoid multiple calls to
vhost_add_used_and_signal(). Which means much less times of used index
updating and memory barriers.

2% performance improvement were seen on netperf TCP_RR test.

Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 280ee66..8a6dd0d 100644 (file)
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -281,7 +281,7 @@ static void vhost_zerocopy_signal_used(struct vhost_net *net,
 {
        struct vhost_net_virtqueue *nvq =
                container_of(vq, struct vhost_net_virtqueue, vq);
-       int i;
+       int i, add;
        int j = 0;
 
        for (i = nvq->done_idx; i != nvq->upend_idx; i = (i + 1) % UIO_MAXIOV) {
@@ -289,14 +289,17 @@ static void vhost_zerocopy_signal_used(struct vhost_net *net,
                        vhost_net_tx_err(net);
                if (VHOST_DMA_IS_DONE(vq->heads[i].len)) {
                        vq->heads[i].len = VHOST_DMA_CLEAR_LEN;
-                       vhost_add_used_and_signal(vq->dev, vq,
-                                                 vq->heads[i].id, 0);
                        ++j;
                } else
                        break;
        }
-       if (j)
-               nvq->done_idx = i;
+       while (j) {
+               add = min(UIO_MAXIOV - nvq->done_idx, j);
+               vhost_add_used_and_signal_n(vq->dev, vq,
+                                           &vq->heads[nvq->done_idx], add);
+               nvq->done_idx = (nvq->done_idx + add) % UIO_MAXIOV;
+               j -= add;
+       }
 }
 
 static void vhost_zerocopy_callback(struct ubuf_info *ubuf, bool success)