From mboxrd@z Thu Jan  1 00:00:00 1970
From: David Ahern <dsa@cumulusnetworks.com>
Subject: Re: [PATCH net-next 2/3] net: mpls: Fixups for GSO
Date: Mon, 26 Sep 2016 20:04:06 -0600
Message-ID: <363ba109-1951-1018-060f-25df33c717e0@cumulusnetworks.com>
References: <1471626542-13335-1-git-send-email-dsa@cumulusnetworks.com>
 <1471626542-13335-3-git-send-email-dsa@cumulusnetworks.com>
 <20160822122122.GA8689@penelope.isobedori.kobe.vergenet.net>
 <dcf3df40-3914-37c6-55dc-e5c3de7530b5@cumulusnetworks.com>
 <20160822145134.GA26491@penelope.isobedori.kobe.vergenet.net>
 <3297517c-2889-19fe-5aa4-89e1d14eadca@cumulusnetworks.com>
 <20160824072002.GA27254@penelope.isobedori.kobe.vergenet.net>
 <CAOrHB_BEn2D_iUgkqPD8ixhjJtnaLpYkANxL+Au7wO=8oKmGag@mail.gmail.com>
 <586321df-72c6-0dee-4ce6-22ca2a0860fb@cumulusnetworks.com>
 <20160926175622.3b00d478@griffin> <20160926190259.3b878aa5@griffin>
Mime-Version: 1.0
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 7bit
Cc: pravin shelar <pshelar@ovn.org>,
        Simon Horman <simon.horman@netronome.com>,
        Pravin B Shelar <pshelar@nicira.com>,
        Linux Kernel Network Developers <netdev@vger.kernel.org>,
        "David S. Miller" <davem@davemloft.net>, buytenh@wantstofly.org,
        "Eric W. Biederman" <ebiederm@xmission.com>, rshearma@brocade.com,
        tom@herbertland.com, Thomas Graf <tgraf@suug.ch>,
        olivier.dugeon@orange.com,
        Alexander Duyck <alexander.duyck@gmail.com>,
        roopa@cumulusnetworks.com
To: Jiri Benc <jbenc@redhat.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-pf0-f180.google.com ([209.85.192.180]:34950 "EHLO
        mail-pf0-f180.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S934017AbcI0CEJ (ORCPT
        <rfc822;netdev@vger.kernel.org>); Mon, 26 Sep 2016 22:04:09 -0400
Received: by mail-pf0-f180.google.com with SMTP id s13so37279128pfd.2
        for <netdev@vger.kernel.org>; Mon, 26 Sep 2016 19:04:09 -0700 (PDT)
In-Reply-To: <20160926190259.3b878aa5@griffin>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On 9/26/16 11:02 AM, Jiri Benc wrote:
> On Mon, 26 Sep 2016 17:56:22 +0200, Jiri Benc wrote:
>> After push_mpls, network_header points to the start of MPLS headers.
>> Which I understand was the point of this patch. However, push_mpls also
>> calls invalidate_flow_key. Meaning that, depending on actions, we may
>> end up calling key_extract soon after. And key_extract sets the network
>> header *after* the MPLS headers.

you know this code better than me, but key_extract pulls the eth header and then sets network header. If MPLS labels are present then it is the labels that the network_header now points to. How did come to the conclusion it is after the labels?

>>
>> That means that on output, for otherwise identical packet,
>> network_header can point before or after MPLS headers based on what
>> actions happened to be executed (recirculation, mainly).
>>
>> If I'm not misreading the code or missing something, this can't be
>> right.
>>
>> mpls_gso_segment does not care, it resets the network_header anyway.
>> What about drivers? What is the correct behavior?
> 
> Answering to myself: it breaks skb_mac_gso_segment. Seems we need to
> fix key_extract to set network_header to the beginning of MPLS headers.
> I'll prepare a patch.
> 
>  Jiri
>