From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1762782AbXGMHXQ (ORCPT ); Fri, 13 Jul 2007 03:23:16 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755545AbXGMHXA (ORCPT ); Fri, 13 Jul 2007 03:23:00 -0400 Received: from mga01.intel.com ([192.55.52.88]:27125 "EHLO mga01.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754119AbXGMHW7 (ORCPT ); Fri, 13 Jul 2007 03:22:59 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.16,536,1175497200"; d="scan'208";a="267783423" Subject: Re: [PATCH][ACPI][BUTTON] remove procfs-interface From: Zhang Rui To: Satyam Sharma Cc: Henne , Arjan van de Ven , "Brown, Len" , linux-acpi@vger.kernel.org, akpm@linux-foundation.org, linux-kernel@vger.kernel.org In-Reply-To: References: <4695F726.7030203@nachtwindheim.de> <823A93EED437D048963A3697DB0E35DE7314AB@pdsmsx414.ccr.corp.intel.com> Content-Type: text/plain Date: Fri, 13 Jul 2007 15:23:51 +0800 Message-Id: <1184311431.4427.12.camel@rzhang-crestline.sh.intel.com> Mime-Version: 1.0 X-Mailer: Evolution 2.6.0 (2.6.0-1) Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 13 Jul 2007 07:22:51.0059 (UTC) FILETIME=[9EFF9830:01C7C51E] Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 2007-07-13 at 03:07 +0800, Satyam Sharma wrote: > On 7/12/07, Zhang, Rui wrote: > > Well, the ACPI sysfs conversion is not finished yet > > [...] > > I'm not sure if the button sysfs I/F is already finished. > > We'd better make a double check. :) > > Ok, this sounds reasonable. > > > and some user space tools still use the ACPI procfs. > > But this does *not*, IMHO. It quite defeats the whole concept of > feature-removal-schedule.txt. I think that file exists precisely > because we cannot gratuitously break userspace interfaces just > like that, but when something gets put up there with a removal date > that is a good one year in the future, and userspace tools _still_ > continue to use it ... then, I suspect something's seriously wrong. > Hi, Satyam, Here I mean the sysfs conversion is not finished, like some ACPI device/driver attributes. i.e. we don't have the alternative in sysfs for all the ACPI proc I/F, which means that part of the ACPI proc I/F are still needed. > Either the feature-removal-schedule.txt file has become something > that users don't even bother checking, or else, they _know_ that > even if they don't bother keeping up with the pace in kernel-land, > that interface still won't go away (because they're still using it!). > In both the above cases, it appears that file itself has become > irrelevant and a "feature" that could be "removed" ... :-) Thanks, Rui