netdev - Re: [PATCH 4/4] udp_tunnel: Pass UDP socket down through udp_tunnel{,6}_xmit

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <1428505382.873082.250792005.6B704B17@webmail.messagingengine.com>
Date:	Wed, 08 Apr 2015 17:03:02 +0200
From:	Hannes Frederic Sowa <hannes@...essinduktion.org>
To:	David Miller <davem@...emloft.net>
Cc:	rick.jones2@...com, tom@...bertland.com, netdev@...r.kernel.org,
	netfilter-devel@...r.kernel.org, pablo@...filter.org,
	jiri@...nulli.us
Subject: Re: [PATCH 4/4] udp_tunnel: Pass UDP socket down through
 udp_tunnel{,6}_xmit_skb().

On Wed, Apr 8, 2015, at 04:13, David Miller wrote:
> From: Hannes Frederic Sowa <hannes@...essinduktion.org>
> Date: Wed, 08 Apr 2015 01:24:59 +0200
> 
> > There is one small problem which already was present before this patch
> > because of my change b6a7719aedd7e5 ("ipv4: hash net ptr into
> > fragmentation bucket selection"). It uses sock_net in iptunnel_xmit,
> > which causes a NULL ptr panic. This patchset did make it more visible
> > because of the explicit NULL passing but the bug was introduced by my
> > patch.
> > 
> > Just wanted to let you know if reports are coming in - I fix it up
> > tomorrow. I will also explore the possibility to give each tunnel at
> > least a raw socket, what Tom proposed.
> > 
> > Rick Jones identified this, thanks!
> 
> Looking into this I also notice that ipip6_tunnel_xmit() passes
> in skb->sk, that needs to be NULL too.
> 
> Want to take care of that along with your sock_net() fix?

Yep, did so and used the simple solution for now.

I am currently trying to check if it makes sense to remove ignore_df and
always use the socket context.

Bye,
Hannes
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html