From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH RFC 1/3] tun: ethtool stats Date: Sat, 07 Oct 2017 00:26:42 +0100 (WEST) Message-ID: <20171007.002642.1079626333365040708.davem@davemloft.net> References: <20171006153059.193688a5@xeon-e3> <20171006161245.0f4abb99@xeon-e3> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: willemdebruijn.kernel@gmail.com, netdev@vger.kernel.org, mst@redhat.com, jasowang@redhat.com, willemb@google.com To: stephen@networkplumber.org Return-path: Received: from shards.monkeyblade.net ([184.105.139.130]:49624 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752278AbdJFX0z (ORCPT ); Fri, 6 Oct 2017 19:26:55 -0400 In-Reply-To: <20171006161245.0f4abb99@xeon-e3> Sender: netdev-owner@vger.kernel.org List-ID: From: Stephen Hemminger Date: Fri, 6 Oct 2017 16:12:45 -0700 > On Fri, 6 Oct 2017 18:37:01 -0400 > Willem de Bruijn wrote: > >> > It looks like you are just exporting the statistics taht are already >> > through normal path. The purpose of ethtool stats is to provide >> > statistics unique to the device, not to repeat what is available throug >> > ip, ifconfig, etc. >> >> A follow-up patch adds non-standard statistics. >> >> I included these common ones, because the new zerocopy counters >> are only interesting in relation to tx_packets. >> >> It also seems customary among existing network drivers to include >> at least these base counters in the ethtool output. > > Only some drivers that slipped through before I started noticing. > Do Not Repeat. Agreed, that is a bad practice unless it is being done to export per-queue versions of those standard netdev statistics.