netdev - Re: [PATCH net] ip_tunnel: Fix dst ref-count.

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

Date:	Mon, 24 Mar 2014 12:36:41 -0700
From:	Pravin Shelar <pshelar@...ira.com>
To:	David Miller <davem@...emloft.net>
Cc:	netdev <netdev@...r.kernel.org>, lucien xin <lucien.xin@...il.com>
Subject: Re: [PATCH net] ip_tunnel: Fix dst ref-count.

On Mon, Mar 24, 2014 at 12:24 PM, David Miller <davem@...emloft.net> wrote:
> From: Pravin <pshelar@...ira.com>
> Date: Sun, 23 Mar 2014 22:06:36 -0700
>
>> Commit 10ddceb22ba (ip_tunnel:multicast process cause panic due
>> to skb->_skb_refdst NULL pointer) removed dst-drop call from
>> ip-tunnel-recv.
>>
>> Following commit reintroduce dst-drop and fix the original bug by
>> checking loopback packet before releasing dst.
>> Original bug: https://bugzilla.kernel.org/show_bug.cgi?id=70681
>>
>> CC: Xin Long <lucien.xin@...il.com>
>> Signed-off-by: Pravin B Shelar <pshelar@...ira.com>
>
> I don't understand why only the GRE tunnel implementation needs the
> check, whereas ipip and sit seem to have the same code associated
> paths.
>
Multicast support is for GRE. The code is always been protected by
CONFIG_NET_IPGRE_BROADCAST. So I do not understand how this applies to
other tunnel protocol. Can you point me to the code you are worried
about?

> Plus I think this is incredibly sloppy, making the implementations
> have to attend to this detail.  Why the heck do we have a common
> piece of infrastructure if we end up with all of this duplicated
> code anyways?
>
> Put this check somewhere common, so that we are guaranteed that the
> issue is really fixed for all tunnel types, no matter what.
>
> Thanks.
--
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