From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx.bisdn.de (mx.bisdn.de [185.27.182.31]) by dpdk.org (Postfix) with ESMTP id 8C96C8084 for ; Fri, 12 Dec 2014 15:10:30 +0100 (CET) Received: from [172.16.250.156] (unknown [172.16.250.156]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mx.bisdn.de (Postfix) with ESMTPSA id 108D2A09C2; Fri, 12 Dec 2014 15:10:30 +0100 (CET) Message-ID: <548AF755.5070401@bisdn.de> Date: Fri, 12 Dec 2014 15:10:29 +0100 From: Marc Sune User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Icedove/31.2.0 MIME-Version: 1.0 To: Sachin Sharma , dev@dpdk.org References: In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit X-Content-Filtered-By: Mailman/MimeDel 2.1.15 Subject: Re: [dpdk-dev] How to add veth interfaces in dpdk 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: Fri, 12 Dec 2014 14:10:30 -0000 On 12/12/14 14:57, Sachin Sharma wrote: > Hi Marc, > I have limited number of nodes (Linux nodes) in my testbed. So, I wanted to use veth interfaces for creating some of dpdk nodes in my nodes. You said that I can use KNI interfaces for this purpose. I am very new to KNI terminology. Could you please redirect me to a tutorial to create links with KNI interfaces and then adding those KNI interfaces into dpdk? For the original problem, you could use qemu VMs with emulated e1000 NICs which are supported by DPDK. This is how we have some development environments here in BISDN. Of course, this is only valid for functional tests, not performance. The KNI documentation is in a section of the DPDK manual. The API is documented here: http://dpdk.org/doc/intel/dpdk-prog-guide-1.7.0.pdf http://dpdk.org/doc/api/rte__kni_8h.html I don't quite get what you meant by "creating links with KNI". KNI interfaces are a bridge between the kernel and the DPDK, meaning a kernel interface (e.g. kni0) will pop up in the kernel, as a regular interface, and from the DPDK zone you can RX and TX from and to it, in a similar way you do in normal eth devices ( using calls rte_kni_rx_burst, rte_kni_tx_burst). Marc > Thanks & Regards, > Sachin. > On 12/12/14 14:34, Sachin Sharma wrote: > >/ Hi all, > />/ > />/ I have created veth interfaces using command "sudo ip link add veth1 type > />/ veth peer name veth2". However, when I use command "sudo > />/ ./tools/igb_uio_bind.py --force --bind=igb_uio veth1" to add veth into > />/ dpdk. It gives me an error that "Unknown device: veth1. Please specify > />/ device in "bus:slot.func" format". I do not see even veth interfaces using > />/ command "sudo ./tools/igb_uio_bind.py --status". Is there any way that I > />/ can add these interfaces into dpdk? > / > >veth are software interfaces, meaning there is no real NIC behind. > >That's why you cannot bind them to igb_uio. > > >You can use KNI interfaces for communicating a DPDK application with the > >kernel interfaces and viceversa. There is an overhead on doing so, but > >whether this is an appropriate solution or not, depends on your use > >case. What do you plan to do? > > >Marc > > > On Fri, Dec 12, 2014 at 2:34 PM, Sachin Sharma > wrote: > > Hi all, > > I have created veth interfaces using command "sudo ip link add > veth1 type veth peer name veth2". However, when I use command > "sudo ./tools/igb_uio_bind.py --force --bind=igb_uio veth1" to add > veth into dpdk. It gives me an error that "Unknown device: veth1. > Please specify device in "bus:slot.func" format". I do not see > even veth interfaces using command "sudo ./tools/igb_uio_bind.py > --status". Is there any way that I can add these interfaces into dpdk? > > > Regards, > Sachin. >