From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eric Dumazet <edumazet@google.com>
Subject: Re: [Patch net v2] mlx5: fixup checksum for short ethernet frame padding
Date: Wed, 28 Nov 2018 19:49:49 -0800
Message-ID: <CANn89iLkNHpCmpAcWa0TiY3A+5vRLeP=oWrQVUEF=yXifnT5CA@mail.gmail.com>
References: <20181128061013.3885-1-xiyou.wangcong@gmail.com>
 <CANn89iL4KM+VpE2ziHvppXZWhGUz-uuaMYTQPMkfxQooc4OpDw@mail.gmail.com>
 <CAM_iQpX2uidJu5ijU7NAVTHm=kNQ+ppvnABZzpbqV53ZqC0mfQ@mail.gmail.com>
 <CANn89iKDNfUP4VzNMxNwKwO8+gyxZwKBr9nSFAeecripnGPYNw@mail.gmail.com>
 <CAM_iQpWXXM3F28b8XLURCJJQaTVK++P9U+u=dehCbqqKEE8VHA@mail.gmail.com>
 <CANn89iJC6CimzRCTP0gfJMVvh_JVoQ9duXsEMo0MOzoEnU9=1A@mail.gmail.com> <CAM_iQpWbCCSMcqeKVO=VWbvGncnm=A5WjQKi1s6UZZkjY3x81g@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Cc: netdev <netdev@vger.kernel.org>,
        Saeed Mahameed <saeedm@mellanox.com>
To: Cong Wang <xiyou.wangcong@gmail.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-it1-f196.google.com ([209.85.166.196]:38727 "EHLO
        mail-it1-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1727079AbeK2Ox4 (ORCPT
        <rfc822;netdev@vger.kernel.org>); Thu, 29 Nov 2018 09:53:56 -0500
Received: by mail-it1-f196.google.com with SMTP id h65so1503978ith.3
        for <netdev@vger.kernel.org>; Wed, 28 Nov 2018 19:50:01 -0800 (PST)
In-Reply-To: <CAM_iQpWbCCSMcqeKVO=VWbvGncnm=A5WjQKi1s6UZZkjY3x81g@mail.gmail.com>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Wed, Nov 28, 2018 at 7:40 PM Cong Wang <xiyou.wangcong@gmail.com> wrote:
>
> On Wed, Nov 28, 2018 at 4:07 PM Eric Dumazet <edumazet@google.com> wrote:
> >
> > A NIC is supposed to deliver frames, even the ones that 'seem' bad.
>
> A quick test shows this is not the case for mlx5.
>
> With the trafgen script you gave to me, with tot_len==40, the dest host
> could receive all the packets. Changing tot_len to 80, tcpdump could no
> longer see any packet. (Both sender and receiver are mlx5.)
>
> So, packets with tot_len > skb->len are clearly dropped before tcpdump
> could see it, that is likely by mlx5 hardware.

Or a router, or a switch.

Are your two hosts connected back to back ?