Thread (8 messages) flat view 8 messages, 3 authors, 2015-01-27

Re: ipv6: oops in datagram.c line 260

From: Steffen Klassert <steffen.klassert@secunet.com>
Date: 2015-01-07 12:26:11

On Wed, Jan 07, 2015 at 11:45:02AM +0100, Hannes Frederic Sowa wrote:
On Mi, 2015-01-07 at 08:22 +0100, Steffen Klassert wrote:
quoted
On Tue, Jan 06, 2015 at 05:01:13PM +0100, Hannes Frederic Sowa wrote:
quoted
xfrm6_output_finish unconditionally resets skb->protocol so we try to
dispatch to the IPv6 handler, even though tcp just sends an IPv4 packet.
Maybe we better dispatch based on sk->sk_family, this should give
always the right address family of the socket.
The original problem was dealing with IPv4/v6 mapped traffic. Processing
local errors from unconnected UDP sockets which are emitting both IPv4
and IPv6 frames won't play nicely with sk->sk_family I fear.
Good point, unfortunately it is not so easy to fix as I thought.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help