From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Dave Young" Subject: Re: [PATCH 1/3][-mm] add class_reclassify macro Date: Wed, 21 May 2008 10:05:42 +0800 Message-ID: References: <20080520095553.GA3201@darkstar.te-china.tietoenator.com> <20080520030232.fc91b64e.akpm@linux-foundation.org> <20080520103045.2bb0a033.akpm@linux-foundation.org> <20080520173640.GP2638@parisc-linux.org> <20080520122356.63bd0000.akpm@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from ti-out-0910.google.com ([209.85.142.190]:36365 "EHLO ti-out-0910.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756595AbYEUCFq (ORCPT ); Tue, 20 May 2008 22:05:46 -0400 Received: by ti-out-0910.google.com with SMTP id b6so1280542tic.23 for ; Tue, 20 May 2008 19:05:43 -0700 (PDT) In-Reply-To: <20080520122356.63bd0000.akpm@linux-foundation.org> Content-Disposition: inline Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Andrew Morton Cc: Matthew Wilcox , greg@kroah.com, kay.sievers@vrfy.org, linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org On Wed, May 21, 2008 at 3:23 AM, Andrew Morton wrote: > On Tue, 20 May 2008 11:36:41 -0600 > Matthew Wilcox wrote: > >> On Tue, May 20, 2008 at 10:30:45AM -0700, Andrew Morton wrote: >> > Well what are these lockdep warnings? Normally such a warning means that >> > we have a locking bug. I _assume_ that you've determined that the warnings >> > are false-positives? >> >> Andrew, we already discussed this on the thread you started that you >> then ignored ... > > rofl. > > All pertinent information should be in a patch's changelog. Then this > sort of confusion will not occur. My wrong. should do this in advance. > >> > The warning which Mariusz Kozlowski discovered ("Subject: Re: >> > 2.6.26-rc2-mm1: possible circular locking dependency detected") was >> > triggered by the "class semaphore to mutex" conversion and it looks >> > like a real bug to me. Would your patch prevent warnings such as that >> > one from being available to us? >> >> The problem is that you add one type of class which then adds devices >> that are of another class. This is not a bug. My proposal is to give >> each sysfs class its own lock class; Dave's is to only do it for the >> two classes he knows about that do this. > > Well that sounds reasonable. I'm not sure that we should introduce > generic-looking helper infrastructure to do it, however. > > Anyway I'll happily sit back and let you guys and Greg sort this one out ;) >