From mboxrd@z Thu Jan  1 00:00:00 1970
From: Matthew Hall <mhall@mhcomputing.net>
Subject: Re: rte_lpm with larger nexthops or another method?
Date: Fri, 26 Jun 2015 00:01:02 -0700
Message-ID: <20150626070102.GA22448@mhcomputing.net>
References: <CANDrEHnuEwX6WGBFbMiB0C38m6axMRiAe3z=9OcjsZkGbPZSSA@mail.gmail.com>
 <20150622175302.GA15788@mhcomputing.net>
 <20150622235102.41c3619a@uryu.home.lan>
 <20150623063024.GA3458@mhcomputing.net>
 <CANDrEHnh-cbT0ATSuvVjJ2xMyNL+tW+s3HY0f+=FSwpNF0uZNg@mail.gmail.com>
 <20150624041314.GA15524@mhcomputing.net>
 <20150624042859.GA16171@mhcomputing.net>
 <20150624051513.GA18427@mhcomputing.net>
 <CANDrEHkP3t4kcbKeDezOVOnrTmgQ8Y+7CkdXwQgeUKgcW38wuQ@mail.gmail.com>
 <20150624175648.GA9719@mhcomputing.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: "<dev@dpdk.org>" <dev@dpdk.org>
To: Vladimir Medvedkin <medvedkinv@gmail.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mail.mhcomputing.net (master.mhcomputing.net [74.208.46.186])
 by dpdk.org (Postfix) with ESMTP id 6A786C7DA
 for <dev@dpdk.org>; Fri, 26 Jun 2015 09:03:43 +0200 (CEST)
Content-Disposition: inline
In-Reply-To: <20150624175648.GA9719@mhcomputing.net>
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

Hi guys,

So I just finally fixed all the weird unit test failures for the basic-mode. 
But the rte_lpm_lookupx4 will take quite a bit longer because it uses a lot of 
real weird SSE intrinsic functions I never used before in my life.

However pretty soon I should get some kind of results how it will work. Sadly 
it appears to be only 1/2 as fast as the old version in the benchmarks, but I 
suspect this wouldn't matter in my specific app, we shall see...

Matthew.