Re: patch: Action repeat

* jamal [ref] 2005-05-04 09:33

On Wed, 2005-04-05 at 15:28 +0200, Thomas Graf wrote:

quoted

* jamal [ref] 2005-05-04 08:59

quoted

If its scope is local i.e for one device, then not reseting could
confuse the next device that sees it and tries to classify on it.

OK, so we're not talking about a reset in action_exec() but rather
in tc_classify() or enqueue()?

in skb_clone() and friends.
Look at CONFIG_NET_CLS_ACT in net/core/skbuff.c

Yes this solves the case for dummy devices etc but how would
this cause a reset on the way from ingress to egress?