From mboxrd@z Thu Jan 1 00:00:00 1970 From: Zoltan Kiss Subject: Sending and receiving on the same port at the same time, from different threads Date: Fri, 15 Jan 2016 16:54:11 +0000 Message-ID: <56992433.7060206@linaro.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit To: dev@dpdk.org Return-path: Received: from mail-wm0-f43.google.com (mail-wm0-f43.google.com [74.125.82.43]) by dpdk.org (Postfix) with ESMTP id 5C6D4568D for ; Fri, 15 Jan 2016 17:54:11 +0100 (CET) Received: by mail-wm0-f43.google.com with SMTP id f206so28017730wmf.0 for ; Fri, 15 Jan 2016 08:54:11 -0800 (PST) Received: from [172.18.45.59] ([195.11.233.227]) by smtp.googlemail.com with ESMTPSA id q129sm3328523wmd.14.2016.01.15.08.54.10 for (version=TLSv1/SSLv3 cipher=OTHER); Fri, 15 Jan 2016 08:54:10 -0800 (PST) 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" Hi, I've been asked this question, and I realized I'm not sure about the answer. In other words: can you call rte_eth_tx_burst() and rte_eth_tx_burst() on the same port at the same time from different threads? In theory it seems possible, as you still access different queues (an RX and a TX one), and at least taking a glance at ixgbe vector functions, they don't seem to use common resources while doing RX or TX. But I'm not sure that it's generally true, although I always assumed that it should be true. Have anyone seen a device where it wasn't true? Regards, Zoltan