net: Move rx skb_orphan call to where needed
authorHerbert Xu <herbert@gondor.apana.org.au>
Mon, 22 Jun 2009 02:25:25 +0000 (02:25 +0000)
committerDavid S. Miller <davem@davemloft.net>
Tue, 23 Jun 2009 23:36:25 +0000 (16:36 -0700)
commitd55d87fdff8252d0e2f7c28c2d443aee17e9d70f
treea4bfb0a47525672c690b63e85bf3f47f4ab014f7
parentd5fdd6babcfc2b0e6a8da1acf492a69fb54b4c47
net: Move rx skb_orphan call to where needed

In order to get the tun driver to account packets, we need to be
able to receive packets with destructors set.  To be on the safe
side, I added an skb_orphan call for all protocols by default since
some of them (IP in particular) cannot handle receiving packets
destructors properly.

Now it seems that at least one protocol (CAN) expects to be able
to pass skb->sk through the rx path without getting clobbered.

So this patch attempts to fix this properly by moving the skb_orphan
call to where it's actually needed.  In particular, I've added it
to skb_set_owner_[rw] which is what most users of skb->destructor
call.

This is actually an improvement for tun too since it means that
we only give back the amount charged to the socket when the skb
is passed to another socket that will also be charged accordingly.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Tested-by: Oliver Hartkopp <olver@hartkopp.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
include/net/sctp/sctp.h
include/net/sock.h
net/ax25/ax25_in.c
net/core/dev.c
net/irda/af_irda.c
net/irda/ircomm/ircomm_lmp.c