From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [RFC] net: remove busylock Date: Thu, 19 May 2016 11:44:10 -0700 Message-ID: <1463683450.18194.220.camel@edumazet-glaptop3.roam.corp.google.com> References: <1463677716.18194.203.camel@edumazet-glaptop3.roam.corp.google.com> <20160519.111208.1949769629804288508.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, aduyck@mirantis.com To: David Miller Return-path: Received: from mail-pf0-f193.google.com ([209.85.192.193]:33853 "EHLO mail-pf0-f193.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932811AbcESSoN (ORCPT ); Thu, 19 May 2016 14:44:13 -0400 Received: by mail-pf0-f193.google.com with SMTP id 145so8940631pfz.1 for ; Thu, 19 May 2016 11:44:13 -0700 (PDT) In-Reply-To: <20160519.111208.1949769629804288508.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: On Thu, 2016-05-19 at 11:12 -0700, David Miller wrote: > From: Eric Dumazet > Date: Thu, 19 May 2016 10:08:36 -0700 > > > busylock was added at the time we had expensive ticket spinlocks > > > > (commit 79640a4ca6955e3ebdb7038508fa7a0cd7fa5527 ("net: add additional > > lock to qdisc to increase throughput") > > > > Now kernel spinlocks are MCS, this busylock things is no longer > > relevant. It is slowing down things a bit. > ... > > That would be a 8 % increase. > > Presumably only for x86-64 and other platforms that actually are using > the ticket spinlocks. > > For others it would be a regression. Ticket spinlocks are actually gone, these are now MCS. But yes, point taken ;)