sending zero length skb's

[Roman Kagan]

On Tue, May 03, 2005 at 12:30:46PM +0200, Duncan Sands wrote:
> Hi Roman, it seems to me that strange things will happen
> if someone tries to send a skb with length zero.  That's
> because ctrl->len will be zero, which is also the signal
> that you've finished with the skb...

Right, I assumed there's no reason to send a zero length skb out.  Was
that wrong?  I might have misinterpreted the protocol description...

Cheers,
  Roman.