From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eric Dumazet <eric.dumazet@gmail.com>
Subject: Re: [PATCH net-next 02/13] sk_buff: add skb extension infrastructure
Date: Thu, 13 Dec 2018 03:16:25 -0800
Message-ID: <2f1509fa-649f-cfc8-7c41-94cd9979ccae@gmail.com>
References: <20181210145006.19098-3-fw@strlen.de>
 <4a43ca01-d09b-71e8-18e1-9a5707787ae0@gmail.com>
 <20181212184446.gyjwbwoyzhrk7kxw@breakpoint.cc>
 <7906a8d9-c2f9-0883-3d13-5ef38e53f10c@gmail.com>
 <20181212205236.sja4fw4sf5egtkyw@breakpoint.cc>
 <8ae68a38-c981-f317-39b1-1092e7efbeeb@gmail.com>
 <20181213092706.fq4mulrp73r2wpq2@breakpoint.cc>
 <46920c3f-564d-e9ef-f714-22f96239736d@gmail.com>
 <20181213103918.rfh3battqdn7u6b6@breakpoint.cc>
 <aa403653-9d65-b997-414f-29024f7c528c@gmail.com>
 <20181213110347.2lopxrbmseiuum7n@breakpoint.cc>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Cc: netdev@vger.kernel.org, cpaasch@apple.com, peter.krystad@intel.com,
        mathew.j.martineau@linux.intel.com
To: Florian Westphal <fw@strlen.de>,
        Eric Dumazet <eric.dumazet@gmail.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-wm1-f65.google.com ([209.85.128.65]:52395 "EHLO
        mail-wm1-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1728067AbeLMLQa (ORCPT
        <rfc822;netdev@vger.kernel.org>); Thu, 13 Dec 2018 06:16:30 -0500
Received: by mail-wm1-f65.google.com with SMTP id m1so1932485wml.2
        for <netdev@vger.kernel.org>; Thu, 13 Dec 2018 03:16:28 -0800 (PST)
In-Reply-To: <20181213110347.2lopxrbmseiuum7n@breakpoint.cc>
Content-Language: en-US
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>



On 12/13/2018 03:03 AM, Florian Westphal wrote:
> Eric Dumazet <eric.dumazet@gmail.com> wrote:
>>
>>
>> On 12/13/2018 02:39 AM, Florian Westphal wrote:
>>>
>>> Thats whats done in the MPTCP out-of-tree implementation, but I don't
>>> think its needed.
>>>
>>> It could just delete the extension before ->queue_xmit() AFAIU.
>>
>> So, cloning would do an refcount_inc(), and deleting the extension would do an refcount_dec_and_test() ?
>>
>> That is what I called an extra pair of atomic operations.
> 
> If it replaces 1:1 current mptcp skb->private out-of-tree storage, then
> yes.
> 

One day I will write a book on the number of atomic operations done on a TCP sendmsg() system call :/