From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933153Ab0E0JNH (ORCPT ); Thu, 27 May 2010 05:13:07 -0400 Received: from mail.windriver.com ([147.11.1.11]:36439 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757778Ab0E0JNF (ORCPT ); Thu, 27 May 2010 05:13:05 -0400 Date: Thu, 27 May 2010 17:11:26 +0800 From: Yong Zhang To: Steven Rostedt Cc: Andrew Morton , monstr@monstr.eu, linux-kernel@vger.kernel.org, michal.simek@petalogix.com, arnd@arndb.de, john.williams@petalogix.com, tglx@linutronix.de, peter.fritzsche@gmx.de, anton@samba.org, mingo@elte.hu, Frederic Weisbecker Subject: Re: [PATCH] asm-generic: Use raw_local_irq_save/restore instead local_irq_save/restore Message-ID: <20100527091126.GB14906@windriver.com> Reply-To: Yong Zhang References: <1274863724-14906-1-git-send-email-monstr@monstr.eu> <1274863724-14906-2-git-send-email-monstr@monstr.eu> <1274884963.22648.245.camel@gandalf.stny.rr.com> <20100526100805.171b3c48.akpm@linux-foundation.org> <1274893911.22648.246.camel@gandalf.stny.rr.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <1274893911.22648.246.camel@gandalf.stny.rr.com> User-Agent: Mutt/1.5.20 (2009-06-14) X-OriginalArrivalTime: 27 May 2010 09:11:29.0404 (UTC) FILETIME=[978FDBC0:01CAFD7C] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, May 26, 2010 at 01:11:51PM -0400, Steven Rostedt wrote: > On Wed, 2010-05-26 at 10:08 -0700, Andrew Morton wrote: > > > > > If a developer looks at atomic_add_return() and asks himself "why did > > this use raw_local_irq_save()", the only way of answering that question > > is to go groveling through the git logs, which is a right PITA if > > you're trying to get some coding work done. > > > > Guys, any time you add code which is non-obvious at the raw C level, it > > *needs* a comment! > > Totally agree! I've gotten pretty good at adding comments to changes > like this that I do. I need to get better at telling others to comment > their work ;-) Cc'ed Frederic Weisbecker Seems like this issue has been raised before: http://marc.info/?l=linux-kernel&m=126292794806169&w=2 I'm afraid this patch will trigger lockdep warning if atomic_inc() is used in some locks.