From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <mhall@mhcomputing.net>
Received: from mail.mhcomputing.net (master.mhcomputing.net [74.208.228.170])
 by dpdk.org (Postfix) with ESMTP id 64E5A6A73
 for <users@dpdk.org>; Tue,  4 Oct 2016 23:35:09 +0200 (CEST)
Received: by mail.mhcomputing.net (Postfix, from userid 1000)
 id EBA0E874; Tue,  4 Oct 2016 14:35:08 -0700 (PDT)
Date: Tue, 4 Oct 2016 14:35:08 -0700
From: Matthew Hall <mhall@mhcomputing.net>
To: Cody Doucette <doucette@bu.edu>
Cc: users@dpdk.org
Message-ID: <20161004213508.GA23228@mhcomputing.net>
References: <CAJjX64YQGKNCamb5HMq_KjCXd6BUX4E6c-8E+QtAhN4DmazOEQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAJjX64YQGKNCamb5HMq_KjCXd6BUX4E6c-8E+QtAhN4DmazOEQ@mail.gmail.com>
User-Agent: Mutt/1.5.21 (2010-09-15)
Subject: Re: [dpdk-users] Subnet filters in hardware
X-BeenThere: users@dpdk.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: usage discussions <users.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/users>,
 <mailto:users-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/users/>
List-Post: <mailto:users@dpdk.org>
List-Help: <mailto:users-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/users>,
 <mailto:users-request@dpdk.org?subject=subscribe>
X-List-Received-Date: Tue, 04 Oct 2016 21:35:09 -0000

On Mon, Oct 03, 2016 at 10:46:07PM -0400, Cody Doucette wrote:
> I'm interested in applying a hardware filter for entire subnets instead of
> individual flows. I've seen previous threads stating that ntuple filter in
> theory may be able to support this, but currently there's not any hardware
> support for it. For example, the 82599 only allows you to set an entire
> field to be matched, but not apply a prefix mask.
> 
> Is anyone aware of any new NICs (last 1-2 years) where this is possible? It
> would be a very useful feature and I'm having a hard time finding anything
> suitable.
> 
> Thanks in advance!
> Cody

It is already possible to achieve line rate subnet matching performance using 
a proper configuration of rte_lpm so this might not really be required. Can 
you try that first?

Matthew.