On Mo, 2014-08-25 at 16:34 -0700, David Miller wrote:

Given Jesper's performance numbers, it's not the way to go.

Instead, go with a signalling scheme via new boolean skb->xmit_more.

This has several advantages:

1) Nearly trivial driver support, just protect the tail pointer
   update with the skb->xmit_more check.

One thing one should keep in mind is, that there must be a skb available
to trigger the flush, maybe this will hurt us one day.

Thinking more about it should we go with a coccinelle script and
replace/extend ndo_start_xmit with an additional argument?

We can also add a new function pointer and call that instead of
ndo_start_xmit. I think only the callq *%rax hurts performance.

Bye,
Hannes