netdev - Re: VETH & AF_PACKET problem

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 for Android: free password hash cracker in your pocket

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

Message-ID: <CACeb84vBLQXA8OMC7RzL=d0-8cYbx5LgojBfTk3vs9dtVs1s0g@mail.gmail.com>
Date:   Tue, 13 Nov 2018 16:54:47 +0530
From:   "Anand H. Krishnan" <anandhkrishnan@...il.com>
To:     netdev@...r.kernel.org
Subject: Re: VETH & AF_PACKET problem

[bump]

Hello,

Any ideas? AF_PACKET with veth seems to be incompatible. I am using DPDK to bind
af_packet with one end of VETH pair and the other end of the pair is
in a container.

Thanks,
Anand
On Mon, Nov 12, 2018 at 12:42 PM Anand H. Krishnan
<anandhkrishnan@...il.com> wrote:
>
> Hello,
>
> We are seeing a problem with AF_PACKET when used along with the
> veth interfaces. SCP complains that message authentication code is
> incorrect.
>
> I was browsing the code and I see that veth_xmit calls ____dev_forward_skb
> which does a skb_scrub_packet, which in turn calls the skb destructor function.
>
> In the case of packets coming from the AF_PACKET socket, the destructor
> function seems to set all the mmap-ed pages to be available for user space to
> copy any new packet it wants. Isn't this a problem?
>
> skb_orphan_frags, called by ____dev_forward_skb, seems to do the right thing,
> but it probably does not get called for packets from AF_PACKET socket, since the
> skb is not a zero copy skb (SKBTX_DEV_ZEROCOPY is not set).
>
> Did I miss something basic here?
>
> (Please cc me, since I am not part of this list)
>
> Thanks,
> Anand