From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264145AbUAUUlF (ORCPT ); Wed, 21 Jan 2004 15:41:05 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264147AbUAUUlF (ORCPT ); Wed, 21 Jan 2004 15:41:05 -0500 Received: from mail.gmx.net ([213.165.64.20]:31654 "HELO mail.gmx.net") by vger.kernel.org with SMTP id S264145AbUAUUlB (ORCPT ); Wed, 21 Jan 2004 15:41:01 -0500 X-Authenticated: #20450766 Date: Wed, 21 Jan 2004 21:37:02 +0100 (CET) From: Guennadi Liakhovetski To: Linus Torvalds cc: "Justin T. Gibbs" , James Bottomley , Linux Kernel Subject: Re: AIC7xxx kernel problem with 2.4.2[234] kernels In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 19 Jan 2004, Linus Torvalds wrote: > On Mon, 19 Jan 2004, Justin T. Gibbs wrote: > > > > Does the maintainer have the ability to veto changes that harm the > > code they maintain? > > Nope. Nobody has that right. > > Even _I_ don't veto changes that the right people push (my motto: > "everybody is wrong sometimes: when enough people complain, even I am > wrong"). > > In particular, maintainers of "conceptually higher" generally always have > priority. If Al Viro says a filesystem is doing something wrong from a VFS > standpoint, then that filesystem is broken - regardless of whether the > filesystem maintainer agrees or not. Because the VFS layer requirements > trump any low-level filesystem issues. Linus May I try to sweeten the pill a bit? I think, I am not contradicting what you said, but just complementing it, thinking, that the direct code maintainer has a right and priority in modifying the code, even over the "conceptionally higher" level. Say, if some code is found to be broken, the problem and possible fixes should first be reported to the direct maintainer. And only if the maintainer is not co-operating suitably (e.g., in the opinion of those "higher" ones), only then necessary modifications can be made directly. In other words, a situation, when, say, a subsystem maintainer silently modifies some driver-code, without even letting the direct maintainer know, is undesirable. A better solution would be to inform the driver maintainer of the problem / send a patch. And only if no suitable action follows, force the necessary modifications. That was just a mere speculation, not pertaining to any specific case. Thanks Guennadi --- Guennadi Liakhovetski