Return-path: Received: from mail.candelatech.com ([208.74.158.172]:37316 "EHLO ns3.lanforge.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1765431Ab3DJSKK (ORCPT ); Wed, 10 Apr 2013 14:10:10 -0400 Message-ID: <5165AB00.2000605@candelatech.com> (sfid-20130410_201038_720666_B1B6CE8C) Date: Wed, 10 Apr 2013 11:10:08 -0700 From: Ben Greear MIME-Version: 1.0 To: "Luis R. Rodriguez" CC: linux-wireless , "linux-kernel@vger.kernel.org" , "netdev@vger.kernel.org" Subject: Re: Jitter and latency benchmarks with netlink / nl80211 References: <5165A874.50804@candelatech.com> In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Sender: linux-wireless-owner@vger.kernel.org List-ID: On 04/10/2013 11:05 AM, Luis R. Rodriguez wrote: > On Wed, Apr 10, 2013 at 10:59 AM, Ben Greear wrote: >> On 04/10/2013 10:55 AM, Luis R. Rodriguez wrote: >>> >>> Curious if anyone has worked on latency and jitter benchmarks in using >>> netlink, specifically with nl80211. Has anyone benchmarked this? Ben, >>> have you? If not for nl80211 perhaps this has been done before for >>> other netlink families? >> >> >> You mean timing related to netlink configuration/report requests? > > Yes. No, I haven't bothered to do this... The way we use netlink is mostly asynchronous..ie we just listen for events and process them as they come in. Most of our configuration would be driven through 'iw', 'wpa_supplicant', 'ip', etc, and much of our stats gathering is through the ethool get-stats API. I would imagine any latency would be highly dependent on other applications taking RTNL (for instance, creating/deleting lots of virtual stations, mac-vlans, ip-address changes, etc). Thanks, Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com