From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <simon.kagstrom@netinsight.net>
Received: from ernst.netinsight.se (ernst.netinsight.se [194.16.221.21])
 by dpdk.org (Postfix) with SMTP id B9C4558D8
 for <dev@dpdk.org>; Mon,  7 Sep 2015 12:41:05 +0200 (CEST)
Received: from [10.100.1.152] (unverified [10.100.1.152]) by
 ernst.netinsight.se
 (EMWAC SMTPRS 0.83) with SMTP id <B0030897279@ernst.netinsight.se>;
 Mon, 07 Sep 2015 12:40:58 +0200
Message-ID: <55ED69BA.4010803@netinsight.net>
Date: Mon, 07 Sep 2015 12:40:58 +0200
From: =?windows-1252?Q?Simon_K=E5gstr=F6m?=
 <simon.kagstrom@netinsight.net>
User-Agent: Mozilla/5.0 (X11; Linux x86_64;
 rv:31.0) Gecko/20100101 Thunderbird/31.8.0
MIME-Version: 1.0
To: Olivier MATZ <olivier.matz@6wind.com>, 
 "Ananyev, Konstantin" <konstantin.ananyev@intel.com>,
 "dev@dpdk.org" <dev@dpdk.org>, "Zhang, Helin" <helin.zhang@intel.com>, 
 "Gonzalez Monroy, Sergio" <sergio.gonzalez.monroy@intel.com>,
 "Burakov, Anatoly" <anatoly.burakov@intel.com>
References: <20150831144110.4a7afa27@miho> <55ED3D9A.7070607@6wind.com>
 <2601191342CEEE43887BDE71AB97725836A83CBA@irsmsx105.ger.corp.intel.com>
 <55ED5A6A.1000803@6wind.com>
In-Reply-To: <55ED5A6A.1000803@6wind.com>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 7bit
Subject: Re: [dpdk-dev] [PATCH RFC] mbuf/ip_frag: Move mbuf chaining to
 common code
X-BeenThere: dev@dpdk.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
X-List-Received-Date: Mon, 07 Sep 2015 10:41:06 -0000

On 2015-09-07 11:35, Olivier MATZ wrote:

>> Wonder why do we need to do that?
>> Probably head mbuf is out of space and want to expand it using pktmbuf_chain()?
>> So in that case seems logical:
>> 1) allocate new mbuf (it's pkt_len will be 0)
>> b) call pktmbuf_chain()
> 
> By experience, having empty segments in the middle of a mbuf
> chain is problematic (functions getting ptr at offsets, some pmds
> or hardware may behave badly), I wanted to avoid that risk.
> 
> Now, the use-case you described is legitimate. Another option would
> be to have another function pktmbuf_append_new(m) that returns a new
> mbuf that is already chained to the other.

I see with that method in that you have to remember to actually update
pkt_len in the head buffer when chaining an empty mbuf. Anyway, to
disallow this behavior should probably not be the responsibility of
rte_pktmbuf_chain(), so I'm fine with leaving the check out.

// Simon