From: Raghuram Kothakota

Sorry I am late in providing my comments, but I feel it is important
to share this comment.

A comment on the original patch...

...

quoted

@@ -811,14 +930,27 @@ static int vnet_start_xmit(struct sk_buff *skb, struct net_device *dev)

	d = vio_dring_cur(dr);

-	tx_buf = port->tx_bufs[dr->prod].buf;
-	skb_copy_from_linear_data(skb, tx_buf + VNET_PACKET_SKIP, skb->len);
+	txi = dr->prod;
+
+	freeskbs = vnet_clean_tx_ring(port, &pending);
+
+	BUG_ON(port->tx_bufs[txi].skb);

	len = skb->len;
-	if (len < ETH_ZLEN) {
+	if (len < ETH_ZLEN)
		len = ETH_ZLEN;
-		memset(tx_buf+VNET_PACKET_SKIP+skb->len, 0, len - skb->len);
+

Aren't you transmitting 'random' bytes from the end of the data?
Plausibly they might not even be mapped.

Also, for short frames the copy could well be faster - especially on systems
with non-trivial iommu.
It is also worth checking whether the original copy was aligned.

	David