From mboxrd@z Thu Jan 1 00:00:00 1970 From: Finn Christensen Subject: Re: [PATCH] ntnic: add PMD driver Date: Mon, 29 Aug 2016 12:00:08 +0000 Message-ID: <02bbac6ecea04602a8ae8dce82d6dc96@napatech.com> References: <20160826134401.19337-1-fc@napatech.com> <20160826095409.6d49a24a@xeon-e3> <8cec6b8c3d6c4a7f85ac117d2d90fd7e@napatech.com> <5255275.7GO0ucYScp@xps13> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Cc: "dev@dpdk.org" , Stephen Hemminger To: Thomas Monjalon Return-path: Received: from mail02.napatech.com (mail02.napatech.com [91.102.88.21]) by dpdk.org (Postfix) with ESMTP id 5248111C5 for ; Mon, 29 Aug 2016 14:00:13 +0200 (CEST) In-Reply-To: <5255275.7GO0ucYScp@xps13> Content-Language: en-US List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" > From: Thomas Monjalon [mailto:thomas.monjalon@6wind.com] > Sent: 29. august 2016 12:04 > To: Finn Christensen > Cc: dev@dpdk.org; Stephen Hemminger > Subject: Re: [dpdk-dev] [PATCH] ntnic: add PMD driver > > 2016-08-29 06:22, Finn Christensen: > > From: Stephen Hemminger [mailto:stephen@networkplumber.org] > > > What is the license of this driver suite? > > > > The driver suite is a closed-source driver, which is not free downloada= ble. > > > > > IMHO the upstream DPDK shouldn't be a platform for non-free driver > suites. > > > > This is our first steps towards opensource, and our upcoming NIC is > > partly build on HW and SW from our Accelerator products. > > And since we already bypasses the kernel effectively in our driver, > > this first solution for a DPDK PMD driver, has been built on top of > > that software suite. > > We like the idea of opensource, but we will need to do the transition > > stepwise, considering our NIC product. > > I think the first step should be to free the lowest level, here the code = you > build your drivers on. I think that will not become a problem. > > We have seen large performance improvements (x4-x7 times with 64 byte > > packets compared to a std NIC in a phy-ovs-vm-ovs-phy setup utilizing > > a modified DPDK), and this is the main motivation to go forward and > > try to push our contribution to DPDK upstream. > > This is the first step of contributions that we want to make. > > This DPDK PMD solution is not compileable unless you have our driver. > > Not being able to compile the PMD is a real problem for maintenance. > The PMD would be considered as dead code, so is forbidden. Yes, that does fully make sense. We will come up with a way to freely downl= oad a driver with header files and libraries to make it possible for anyone to bu= ild the PMD. > > > We may need to make that possible, so that a free downloadable driver > > can be retrieved. > > > > Once this is said, we thought that the DPDK was BSD licensed and I > > must admit that we have failed to see this limitation that you are > mentioning. > > Is there another license or agreement text that we need to read? > > Maybe you'll find some interesting parts in the contributing guide. > But honestly, I think you have already done the right thing in your case: > you sent some code and open the discussion :) Now you just need to enable > a free compilation environment for your patch. > Thanks We will do that. I'll get back with an updated patch once this has been est= ablished. Thank you both for your help on this. Disclaimer: This email and any files transmitted with it may contain confid= ential information intended for the addressee(s) only. The information is n= ot to be surrendered or copied to unauthorized persons. If you have receive= d this communication in error, please notify the sender immediately and del= ete this e-mail from your system.