From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755471Ab1HBWG2 (ORCPT ); Tue, 2 Aug 2011 18:06:28 -0400 Received: from one.firstfloor.org ([213.235.205.2]:50913 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755390Ab1HBWGX (ORCPT ); Tue, 2 Aug 2011 18:06:23 -0400 Date: Wed, 3 Aug 2011 00:06:19 +0200 From: Andi Kleen To: David Rientjes Cc: Alan Cox , Linus Torvalds , Andrew Morton , Andi Kleen , linux-kernel@vger.kernel.org Subject: Re: Revert needed: udev spewing warnons on common systems in 3.0 Message-ID: <20110802220619.GI27083@one.firstfloor.org> References: <20110802005206.GA3832@tassilo.jf.intel.com> <20110801183207.cea89a96.akpm@linux-foundation.org> <20110802101614.009cdf42@lxorguk.ukuu.org.uk> <20110802111117.10a73b89@lxorguk.ukuu.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.2i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > All of the examples that I've cited that were fixed as a result of the > printk_once() were through bugzilla for the developers of those > applications, I didn't talk to any distro when I publically said I would > handle all reports of this warning that people didn't know how to triage. The distros would have updated udev at some point anyways. But they won't do it for old releases. But the users who don't want to update will just be annoyed (in Linus words) the backtrace forever. The big problem I have with your "use kernel WARN_ONs to social engineer" approach is that the devalues a good mechanism to report kernel problems. Traditionally at least I looked at all backtraces and tried to get at the bottom of the problem to improve the kernel. I know from email that users reported those too. Now that all my test systems spew them on every boot I cannot do that anymore. When I see a backtrace scrolling by I now just think "Ah it's probably David Rientjes trying to social engineer someone again". And then some point noone will look at backtraces anymore because they have become normal. Really serious problems will get missed. And this is a bad thing. IMHO still the only sensible thing that can be done with this patch is to revert it. -Andi