From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH net-next 1/6] net: sched: sch_api: handle generic qdisc errors Date: Thu, 07 Dec 2017 13:08:15 -0500 (EST) Message-ID: <20171207.130815.2195718476285159124.davem@davemloft.net> References: <75db5547-08bb-c3cb-ef7b-f3f73f66709d@gmail.com> <61d9b390-3193-3e17-ea65-30f457647eae@mojatatu.com> <5b8015b8-4fd2-0b1e-2e78-9b318b529f30@gmail.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: jhs@mojatatu.com, aring@mojatatu.com, xiyou.wangcong@gmail.com, jiri@resnulli.us, netdev@vger.kernel.org, kernel@mojatatu.com To: dsahern@gmail.com Return-path: Received: from shards.monkeyblade.net ([184.105.139.130]:41506 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754470AbdLGSIR (ORCPT ); Thu, 7 Dec 2017 13:08:17 -0500 In-Reply-To: <5b8015b8-4fd2-0b1e-2e78-9b318b529f30@gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: From: David Ahern Date: Thu, 7 Dec 2017 10:52:01 -0700 > What value are the messages providing above and beyond the standard libc > strerror(errno)? In the case of ENOMEM, there is nothing in the user can > do to fix that particular command, so why bloat the code with extraneous > messages? Similarly with other errors like ENODEV -- if there is only 1 > device in question AND the user specified the device then you do not > need to augment with an additional error message. Agreed, all of this stuff is bloat and not useful in any way to the user.