From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: Broken RSS hash computation on Intel 82574L Date: Tue, 1 Sep 2015 17:50:16 +0300 Message-ID: <55E5BB28.1090009@cloudius-systems.com> References: <55E44563.5050208@ics.muni.cz> <55E5B81E.8010601@ics.muni.cz> <20150901144726.GA30421@mhcomputing.net> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: quoted-printable Cc: dev@dpdk.org To: Matthew Hall , =?UTF-8?Q?Martin_Dra=c5=a1ar?= Return-path: Received: from mail-wi0-f176.google.com (mail-wi0-f176.google.com [209.85.212.176]) by dpdk.org (Postfix) with ESMTP id 16C0A568C for ; Tue, 1 Sep 2015 16:50:19 +0200 (CEST) Received: by wiclp12 with SMTP id lp12so33634598wic.1 for ; Tue, 01 Sep 2015 07:50:19 -0700 (PDT) In-Reply-To: <20150901144726.GA30421@mhcomputing.net> 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" On 09/01/2015 05:47 PM, Matthew Hall wrote: > On Tue, Sep 01, 2015 at 04:37:18PM +0200, Martin Dra=C5=A1ar wrote: >> Dne 1.9.2015 v 15:45 De Lara Guarch, Pablo napsal(a): >>> 82574L NIC uses em PMD, which does not support more than 1 queue. >>> Therefore RSS is disabled in the NIC and then you cannot have RSS has= hes. >>> >>> Thanks, >>> Pablo >> Hi Pablo, >> >> that is an interesting information. I read the rationale in em_ethdev.= c >> and I was wondering, what would have to be done to enable RSS hash >> computation on that card. I can live with just one RX queue, but hashe= s >> would help me a lot. The computer which is using those NICs is not tha= t >> powerful and every bit of offloaded computation counts... >> >> Thanks, >> Martin > RSS calculations are used to direct packets across multiple RX queues. = With > only one RX queue it cannot possibly increase performance by enabling i= t. > As an example, seastar uses the RSS hash computed by the NIC to select a=20 core to process on, if the number of hardware queues is smaller than the=20 number of cores.