From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga11.intel.com (mga11.intel.com [192.55.52.93]) by dpdk.org (Postfix) with ESMTP id AAAE17E75 for ; Thu, 13 Nov 2014 00:04:09 +0100 (CET) Received: from fmsmga001.fm.intel.com ([10.253.24.23]) by fmsmga102.fm.intel.com with ESMTP; 12 Nov 2014 15:14:04 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.07,371,1413270000"; d="scan'208";a="621489972" Received: from irsmsx101.ger.corp.intel.com ([163.33.3.153]) by fmsmga001.fm.intel.com with ESMTP; 12 Nov 2014 15:14:01 -0800 Received: from irsmsx151.ger.corp.intel.com (163.33.192.59) by IRSMSX101.ger.corp.intel.com (163.33.3.153) with Microsoft SMTP Server (TLS) id 14.3.195.1; Wed, 12 Nov 2014 23:14:01 +0000 Received: from irsmsx105.ger.corp.intel.com ([169.254.7.56]) by IRSMSX151.ger.corp.intel.com ([169.254.4.227]) with mapi id 14.03.0195.001; Wed, 12 Nov 2014 23:14:01 +0000 From: "Ananyev, Konstantin" To: Thomas Monjalon , Olivier MATZ , "dev@dpdk.org" Thread-Topic: [dpdk-dev] [PATCH v8 10/10] app/testpmd:test VxLAN Tx checksum offload Thread-Index: AQHP+igwSkdCwquBBkeqJNewvNm7F5xVaLcAgARPPrCAAFHjgIACB2EwgAD3JlqAADp1EA== Date: Wed, 12 Nov 2014 23:14:00 +0000 Message-ID: <2601191342CEEE43887BDE71AB977258213A4216@IRSMSX105.ger.corp.intel.com> References: <1414376006-31402-1-git-send-email-jijiang.liu@intel.com> <2601191342CEEE43887BDE71AB977258213A3F5F@IRSMSX105.ger.corp.intel.com> <54635B2B.5040603@6wind.com> <1893731.yh14ESvpZM@xps13> In-Reply-To: <1893731.yh14ESvpZM@xps13> Accept-Language: en-IE, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [163.33.239.181] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v8 10/10] app/testpmd:test VxLAN Tx checksum offload X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 12 Nov 2014 23:04:10 -0000 Hi Thomas, > -----Original Message----- > From: Thomas Monjalon [mailto:thomas.monjalon@6wind.com] > Sent: Wednesday, November 12, 2014 1:41 PM > To: Olivier MATZ; Ananyev, Konstantin; dev@dpdk.org > Cc: Yong Wang > Subject: Re: [dpdk-dev] [PATCH v8 10/10] app/testpmd:test VxLAN Tx checks= um offload >=20 > 2014-11-12 14:05, Olivier MATZ: > > On 11/12/2014 10:55 AM, Ananyev, Konstantin wrote: > > >> From an API perspective, it looks a bit more complex to have to cal= l > > >> dev_prep_tx() before sending the packets if they have been flagged > > >> for offload processing. But I admit I have no other argument. I'll b= e > > >> happy to have more comments from other people on the list. > > >> > > >> I'm sending a first version of the patchset now as it's ready, it do= es > > >> not take in account this comment, but I'm open to add it in a v2 if > > >> there is a consensus on this. > > >> > > >> Now, knowing that: > > >> - adding dev_prep_tx() will also concern hw checksum (TCP L4 checksu= m > > >> already requires to set the TCP pseudo header checksum), so addin= g > > >> this will change the API of an existing feature > > >> - TSO is a new feature expected for 1.8 (which should be out soon) > > >> > > >> Do you think we need to include this for 1.8 or can we postpone your > > >> proposition for after the 1.8 release? > > > > > > I'd say it would be good to have it done together with TSO feature. > > > About changing API: I think existing applications shouldn't be affect= ed. > > > For existing PMDs/TX offloads we don't change any rules what need to= be filled by the app. > > > We just add a new function that can do that for user. > > > If the app fills required manually (as all apps have to do now) it wo= uld keep working as expected. > > > > I agree, this proposition could work without changing the current > > applications. > > > > > If you feel like it is too much work for 1.8 timeframe - > > > can we at least move fix_tcp_phdr_cksum() out of TX PMD as a temporar= y measure? > > > Let say create a function get_ipv4_udptcp_checksum(struct rte_mbuf *= m) (in librte_net ?). > > > It will calculate PSD checksum for both TSO and non-TSO case based on= given mbuf flags/fields. > > > Then we can update testpmd/csumonly.c to use it. > > > > I'm not sure having get_ipv4_udptcp_checksum() in librte_net would > > help. The value we have to set in the TCP checksum field depends on the > > PMD (altought only ixgbe is supported now). So, it would require > > another parameter and a new PMD eth_ops... which looks very > > similar to dev_prep_tx() (except that dev_prep_tx() can be bulked). > > I think a stack will not be able to call get_udptcp_checksum(m ,port) > > because it does not know the physical port at the time the packet is > > built. Moreover, calling a function through a pointer is more efficient > > when bulked. So I think the dev_prep_tx() you initially describe is > > a better answer to the problem. > > > > I don't know what is the exact timeframe for 1.8, maybe Thomas can help > > on this? Depending on it, we have several options: > > > > - implement dev_prep_tx() for 1.8 in the TSO series: this implies that > > the community agrees on this new API. We need to check that it will > > be faster in a pipeline model (I think this is obvious) but also tha= t > > it does not penalize the run-to-completion model: introducing anothe= r > > function dev_prep_tx() can result in duplicated tests in the driver > > (ex: test the offload flag values). > > > > - postpone dev_prep_tx() or similar to next version and push the curren= t > > TSO patchset (including the comments done on the list). It does not > > modify the current offload API, it provides the TSO feature on ixgbe > > based on a similar API concept (set the TCP phdr cksum). The drawbac= k > > is a potential performance loss when using a pipeline model. > > > > - another option that you may prefer is to bind the API behavior to > > ixgbe (for 1.8): we can ask the application to set the pseudo-header > > checksum without the IP len when doing TSO, as required by the ixgbe > > driver. Then, for next release, we can think about dev_prep_tx(). Th= e > > drawback of this solution is that we may go back on this choice if t= he > > dev_prep_tx() approach is not validated by the community. >=20 > I feel this question is really important and we need more people to revie= w > the API. We'll also need more validation tests and performance checks wit= h > several use cases. >=20 > Release is already late and I'm not comfortable with such change now. > The only chance to have dev_prep_tx() in 1.8 would be to quickly have a l= arge > consensus and some benchmarks in pipeline and run to completion models. >=20 > Conclusion: we should integrate TSO without dev_prep_tx (option 2 or 3) a= nd > then speed up dev & tests for dev_prep_tx(). This improvement for pipelin= e > model could go in 2.0 if it's considered too short or risky for 1.8. > Konstantin, could you be in charge of dev_prep_tx() works? I can have a look at it in 2.0 timeframe. Unless someone else is interested in doing it before that :) Konstantin >=20 > Thanks for the good discussion > -- > Thomas