From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH v4] net: dev_weight: TX/RX orthogonality Date: Thu, 29 Dec 2016 14:08:54 -0500 (EST) Message-ID: <20161229.140854.1456743873101323068.davem@davemloft.net> References: <20161228.141751.81302085672323860.davem@davemloft.net> <1483005521-27799-1-git-send-email-matthias.tafelmeier@gmx.net> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, hagen@jauu.net, fw@strlen.de, edumazet@google.com, daniel@iogearbox.net To: matthias.tafelmeier@gmx.net Return-path: Received: from shards.monkeyblade.net ([184.105.139.130]:56338 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753247AbcL2TI4 (ORCPT ); Thu, 29 Dec 2016 14:08:56 -0500 In-Reply-To: <1483005521-27799-1-git-send-email-matthias.tafelmeier@gmx.net> Sender: netdev-owner@vger.kernel.org List-ID: From: Matthias Tafelmeier Date: Thu, 29 Dec 2016 10:58:41 +0100 > Oftenly, introducing side effects on packet processing on the other half > of the stack by adjusting one of TX/RX via sysctl is not desirable. > There are cases of demand for asymmetric, orthogonal configurability. > > This holds true especially for nodes where RPS for RFS usage on top is > configured and therefore use the 'old dev_weight'. This is quite a > common base configuration setup nowadays, even with NICs of superior processing > support (e.g. aRFS). > > A good example use case are nodes acting as noSQL data bases with a > large number of tiny requests and rather fewer but large packets as responses. > It's affordable to have large budget and rx dev_weights for the > requests. But as a side effect having this large a number on TX > processed in one run can overwhelm drivers. > > This patch therefore introduces an independent configurability via sysctl to > userland. This is missing a signoff.