On Tue, Feb 17, 2009 at 01:53:25PM +0100, Karl Hiramoto wrote:
Jarek Poplawski wrote:
quoted
On Tue, Feb 17, 2009 at 12:49:07PM +0100, Karl Hiramoto wrote:
...
quoted
A side note: so far the original patch i sent works in all cases i have
tested, but fails with tcpdump. I suspect its because the skb gets cloned.
Hmm... I would like to make sure: if tcpdump breaks it too, then
we don't need to blame skb_cow_head(), do we? If it's like this then
it looks like the driver's problem with copied/cloned skbs?! (Maybe
because some "private" offset is overwritten?)
Jarek P.