Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757489AbdIILXO (ORCPT ); Sat, 9 Sep 2017 07:23:14 -0400 Received: from mail-wm0-f50.google.com ([74.125.82.50]:45728 "EHLO mail-wm0-f50.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753471AbdIILXM (ORCPT ); Sat, 9 Sep 2017 07:23:12 -0400 X-Google-Smtp-Source: ADKCNb4nTLFL4/TybK0Wg0/rgbZxOaJLgvusyCvtLJFZpzrzohVFCTNRwFDdVMRqH/Yie4BB19dJsg== Subject: Re: [PATCH] net: bonding: Fix transmit load balancing in balance-alb mode if specified by sysfs To: =?UTF-8?B?TWFoZXNoIEJhbmRld2FyICjgpK7gpLngpYfgpLYg4KSs4KSC4KSh4KWH4KS1?= =?UTF-8?B?4KS+4KSwKQ==?= References: <17EC94B0A072C34B8DCF0D30AD16044A0298684C@BPXM09GP.gisp.nec.co.jp> <03b61877-053b-2f0e-dc35-8fe31cc90c08@cumulusnetworks.com> <9ae82441-78da-0056-88c1-2f9d863fb81f@cumulusnetworks.com> Cc: Kosuke Tatsukawa , Jay Vosburgh , Veaceslav Falico , Andy Gospodarek , "netdev@vger.kernel.org" , "linux-kernel@vger.kernel.org" , Reinis Rozitis From: Nikolay Aleksandrov X-Enigmail-Draft-Status: N1110 Message-ID: <43356b9d-77e5-4ccb-ee1b-f293d2fb130c@cumulusnetworks.com> Date: Sat, 9 Sep 2017 14:23:08 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Icedove/45.6.0 MIME-Version: 1.0 In-Reply-To: <9ae82441-78da-0056-88c1-2f9d863fb81f@cumulusnetworks.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1257 Lines: 30 On 09/09/17 13:29, Nikolay Aleksandrov wrote: > On 09/09/17 02:54, Mahesh Bandewar (महेश बंडेवार) wrote: >> On Fri, Sep 8, 2017 at 7:30 AM, Nikolay Aleksandrov >> wrote: >>> On 08/09/17 17:17, Kosuke Tatsukawa wrote: [snip] >>> >> I think the underlying issue is that tlb_dynamic_lb should be set to 1 >> for all modes which was not the case when it was getting initialized >> only forTLB mode. So from that perspective I prefer Nik's patch with a >> small variation that guards the case when mode transitions from TLB to >> ALB. The reason why I like that patch is because it's simple and >> avoids complications. > > +1, I think this is the most straight-forward solution as well and > safest for -net > > I will go ahead and submit it in a few minutes. > > Thanks, > Nik > Just FYI, since the second fix (tlb_dynamic_lb in TLB = 0 switch to ALB) is identical to this patch, I'm acking this one and will wait until it's in to submit the default value fix (if we start in non-TLB mode and switch to TLB we'll get dynamic_lb = 0 currently). I guess this is the simplest way, I didn't want to alter user-configured value on mode switch, but it seems better than the alternative especially for -net.