[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHA+R7MrL4kW5W=1LZFzXzkb1zWUxHRXHwjjnksScWF1ZMHswQ@mail.gmail.com>
Date: Fri, 20 Mar 2015 13:55:11 -0700
From: Cong Wang <cwang@...pensource.com>
To: Daniel Mack <daniel@...que.org>
Cc: Florian Westphal <fw@...len.de>,
Daniel Borkmann <dborkman@...hat.com>,
Alexey Perevalov <a.perevalov@...sung.com>,
Pablo Neira Ayuso <pablo@...filter.org>,
netdev <netdev@...r.kernel.org>
Subject: Re: cgroup matches in INPUT chain
On Fri, Mar 20, 2015 at 9:21 AM, Daniel Mack <daniel@...que.org> wrote:
> On 03/20/2015 05:11 PM, Florian Westphal wrote:
>> Daniel Mack <daniel@...que.org> wrote:
>>> In my simple test setup, when skbs are dequeued by process_backlog(),
>>> they have skb->_skb_refdst set, and hence ip_rcv_finish() does not call
>>> into early_demux() prior to iterating the INPUT chain:
>>
>> Yes, because we already have a route set.
>>
>> Are we talking about loopback?
>
> I'm testing this on the lookback device, but I've seen similar behavior
> on external interfaces too. However, I fail to see a pattern in that.
>
Loopback is special because the skb->dst is kept across TX and RX.
How possible is your external interface sets dst for the packets?
Are you using a tunnel device or you have some other setup you didn't mention?
--
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
Powered by blists - more mailing lists