From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <sodey@sonusnet.com>
Received: from p01c12o144.mxlogic.net (p01c12o144.mxlogic.net [208.65.145.67])
 by dpdk.org (Postfix) with ESMTP id F3299212
 for <dev@dpdk.org>; Wed, 15 Oct 2014 13:44:55 +0200 (CEST)
Received: from unknown [66.151.187.12] (EHLO smtp.sonusnet.com)
 by p01c12o144.mxlogic.net(mxl_mta-8.2.0-0) over TLS secured channel
 with ESMTP id b006e345.0.46564.00-392.116751.p01c12o144.mxlogic.net
 (envelope-from <sodey@sonusnet.com>); 
 Wed, 15 Oct 2014 05:52:45 -0600 (MDT)
X-MXL-Hash: 543e600d575e83c3-ff3a22a598745b5230ee0f410cadc86786f00a1c
Received: from INBA-HUB02.sonusnet.com (10.70.51.87) by
 psmwsonshc02.sonusnet.com (10.176.20.8) with Microsoft SMTP Server (TLS) id
 14.3.158.1; Wed, 15 Oct 2014 07:52:42 -0400
Received: from INBA-MAIL02.sonusnet.com ([10.70.51.89]) by
 inba-hub02.sonusnet.com ([fe80::80b9:dc60:caf7:7dfc%11]) with mapi id
 14.03.0181.006; Wed, 15 Oct 2014 17:22:40 +0530
From: "Dey, Souvik" <sodey@sonusnet.com>
To: Thomas Monjalon <thomas.monjalon@6wind.com>
Thread-Topic: [dpdk-dev] FW: BUG in IP FRAGMENTATION
Thread-Index: Ac/oZ0B22lKbfrCoTWKqR8L2bLPc9gAAKr1A//+o6YD//6G94A==
Date: Wed, 15 Oct 2014 11:52:39 +0000
Message-ID: <4B8F36DD0FB25E47B3DA6F493BF591B70DC597C1@inba-mail02.sonusnet.com>
References: <4B8F36DD0FB25E47B3DA6F493BF591B70DC5972B@inba-mail02.sonusnet.com>
 <471124659.68G021lYDz@xps13>
In-Reply-To: <471124659.68G021lYDz@xps13>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.70.54.43]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-AnalysisOut: [v=2.1 cv=asamCEpV c=1 sm=1 tr=0 a=OF4WD78bns2OWLoCgEPmtw==]
X-AnalysisOut: [:117 a=OF4WD78bns2OWLoCgEPmtw==:17 a=URNJFqk9g_oA:10 a=4Rb]
X-AnalysisOut: [SuBE0YzsA:10 a=BLceEmwcHowA:10 a=kj9zAlcOel0A:10 a=xqWC_Br]
X-AnalysisOut: [6kY4A:10 a=kUVcWBOSAAAA:8 a=YlVTAMxIAAAA:8 a=4YC7WVefAAAA:]
X-AnalysisOut: [8 a=8rWy6zfcAAAA:8 a=XfjEpyLPcTd-1dCrjboA:9 a=CjuIK1q_8ugA]
X-AnalysisOut: [:10 a=pjrC_YA95a8A:10 a=SumTzdpxxCEA:10]
X-Spam: [F=0.5000000000; CM=0.500; MH=0.500(2014101506); S=0.200(2014051901)]
X-MAIL-FROM: <sodey@sonusnet.com>
X-SOURCE-IP: [66.151.187.12]
Cc: "dev@dpdk.org" <dev@dpdk.org>, "Patil, PraveenKumar" <ppatil@sonusnet.com>
Subject: Re: [dpdk-dev] FW: BUG in IP FRAGMENTATION
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: Wed, 15 Oct 2014 11:44:57 -0000

We have not directly tried to use the 1.7 code with fragmentation apis, but=
 we did run through the 1.7 code and did not find any much difference betwe=
en 1.6 and 1.7 code.=20
I had wrongly mentioned out-of-order in my previous mail. Actually out-of-o=
rder is working fine but we are facing issues with overlapping and duplicat=
e fragments.
 In the 1.7 fragmentation code also in file ip_frag_internals.c , function =
ip_frag_process we also see this comment

	/*
	 * errorneous packet: either exceeed max allowed number of fragments,
	 * or duplicate first/last fragment encountered.
	 */

Which indirectly suggest that the handling of duplicate first/last fragment=
 is taken as error. Same with overlapping fragment we could not find any pi=
ece of code which will be doing it.

--
Regards,
Souvik

-----Original Message-----
From: Thomas Monjalon [mailto:thomas.monjalon@6wind.com]=20
Sent: Wednesday, October 15, 2014 4:54 PM
To: Dey, Souvik
Cc: dev@dpdk.org; Patil, PraveenKumar
Subject: Re: [dpdk-dev] FW: BUG in IP FRAGMENTATION

ME TOO, I HAVE A BUG WITH CAPS LOCK ;)

2014-10-15 11:06, Dey, Souvik:
> In DPDK1.6 do we support overlapped fragments while doing reassembly.
> Also why we don't consider the first or last fragment to be out-of-order.
> Are this known limitations in DPDK or they are not working due to some=20
> bugs in the code ?

Please test the latest version and explain how you see the bug.

Thanks
--
Thomas