From mboxrd@z Thu Jan 1 00:00:00 1970 From: Willy Tarreau Subject: Re: [PATCH 1/1] net: Add SO_REUSEPORT_LISTEN_OFF socket option as drain mode Date: Thu, 24 Mar 2016 19:24:35 +0100 Message-ID: <20160324182435.GA7682@1wt.eu> References: <1458828813.10868.65.camel@edumazet-glaptop3.roam.corp.google.com> <20160324142222.GB7237@1wt.eu> <1458830744.10868.72.camel@edumazet-glaptop3.roam.corp.google.com> <20160324153053.GA7569@1wt.eu> <1458837191.12033.4.camel@edumazet-glaptop3.roam.corp.google.com> <20160324165047.GA7585@1wt.eu> <1458838897.12033.10.camel@edumazet-glaptop3.roam.corp.google.com> <56F42A00.7050002@iogearbox.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Daniel Borkmann , Tom Herbert , Eric Dumazet , Craig Gallek , Josh Snyder , Aaron Conole , "David S. Miller" , Linux Kernel Network Developers To: Tolga Ceylan Return-path: Received: from wtarreau.pck.nerim.net ([62.212.114.60]:32149 "EHLO 1wt.eu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751154AbcCXSYx (ORCPT ); Thu, 24 Mar 2016 14:24:53 -0400 Content-Disposition: inline In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On Thu, Mar 24, 2016 at 11:20:49AM -0700, Tolga Ceylan wrote: > I would appreciate a conceptual description on how this would work > especially for a common scenario > as described by Willy. My initial impression was that a coordinator > (master) process takes this > responsibility to adjust BPF filters as children come and go. Indeed that would help, I don't know where to start from for now. > Two popular software has similar use cases: nginx and haproxy. Another > concern is with the > introduction of BPF itself, should we expect a performance drop in > these applications? Knowing how picky Eric is about performance in such areas, I'm not worried a single second about adopting something he recommends :-) I just need to ensure it covers our users' needs. And maybe the solution I mentionned in the other e-mail could work. Willy