From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758328AbbDWUca (ORCPT ); Thu, 23 Apr 2015 16:32:30 -0400 Received: from mail-wg0-f50.google.com ([74.125.82.50]:32791 "EHLO mail-wg0-f50.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754209AbbDWUc3 (ORCPT ); Thu, 23 Apr 2015 16:32:29 -0400 Date: Thu, 23 Apr 2015 23:32:26 +0300 From: Alexey Dobriyan To: Tejun Heo Cc: xiaoming.wang@intel.com, Linux Kernel , Mel Gorman , Andrew Morton Subject: Re: [PATCH] proc: move the adding option Ngid to the end of proc/PID/status Message-ID: <20150423203226.GA1765@p183.telecom.by> References: <20150417142605.GA1954@htj.duckdns.org> <20150417151259.GA16743@htj.duckdns.org> <20150421140007.GA22502@p183.telecom.by> <20150421151119.GB9455@htj.duckdns.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150421151119.GB9455@htj.duckdns.org> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Apr 21, 2015 at 11:11:19AM -0400, Tejun Heo wrote: > On Tue, Apr 21, 2015 at 05:00:07PM +0300, Alexey Dobriyan wrote: > > > The only reason for changing the position is because > > > there's this specific breakage. The goal should be working around > > > that specific case while keeping the impact minimum on everyone else. > > > > If there are TWO incorrect parsers, one for TracerPid, another for Ngid, > > you CAN'T workaround it. And if you can't workaround you choose code > > which was written first, namely, TracerPid one. > > Not when the code has been out for 1.5 years. Minimizing the > disturbance is the better course of action. Look at the file. If you > move ngid to the end now, it's gonna shift most of the file content, > which is what caused the problem in the first place. > > We don't know what's out there which again was the same problem which > triggered this thread in the first place. Why would you take the same > amount of risk when you can fix the known issue with less amount of > changes? There are 2 fields before Ngid and 35+ after Ngid. So the risk is not the same. Potentionally, Ngid addition broke almost every parser. > Just put ngid after tracerpid. That way, we can fix the > known problems while changing the offsets of only four fields. At > this point, no change to the file layout is "right". Such thing isn't > defined regardless of who came first. The only thing we can do is > working around the known cases while minimizing possible impacts. We'll return to this thread when next breakage will be reported, I promise. :^)