From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933345AbbHZBhz (ORCPT ); Tue, 25 Aug 2015 21:37:55 -0400 Received: from eu-smtp-delivery-143.mimecast.com ([146.101.78.143]:58849 "EHLO eu-smtp-delivery-143.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932758AbbHZBhx convert rfc822-to-8bit (ORCPT ); Tue, 25 Aug 2015 21:37:53 -0400 Date: Wed, 26 Aug 2015 09:37:47 +0800 From: Huang Shijie To: Thomas Gleixner Cc: jiang.liu@linux.intel.com, marc.zyngier@arm.com, linux-kernel@vger.kernel.org Subject: Re: [PATCH] irqdomain: remove the confused log Message-ID: <20150826013747.GA30900@free64-pc.asiapac.arm.com> References: <1440496131-29192-1-git-send-email-shijie.huang@arm.com> MIME-Version: 1.0 In-Reply-To: User-Agent: Mutt/1.5.23 (2014-03-12) X-OriginalArrivalTime: 26 Aug 2015 01:37:50.0299 (UTC) FILETIME=[D198CAB0:01D0DF9F] X-MC-Unique: _cdATohrQzCqc6paLBTG9A-1 Content-Type: text/plain; charset=WINDOWS-1252 Content-Transfer-Encoding: 8BIT Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Aug 25, 2015 at 09:34:00PM +0200, Thomas Gleixner wrote: > On Tue, 25 Aug 2015, Huang Shijie wrote: > > > When the dynamic debug is enabled for irq domain, the kernel will print out: > > ------------------------------------------------------ > > irqdomain:__irq_domain_add: irq: Added domain (null) > > ------------------------------------------------------ > > > > It does not print out the irq domain's name, since the @domain->name is NULL. > > Keep the log here just makes people nervous. > > Well, it might be NULL in your case, but there are enough cases where > its not NULL and the output valuable. In which case ? The domain is allocated by kzalloc_node(), so @domain->name is NULL in the function. The @domain->name is set in other function, such as irq_domain_associate(). > > > Aside of that I don't understand what's making one nervous > here. People who enable that debug stuff should know that this can be > NULL and just deal with it. If they get nervous about it they probably > should not try to debug that code in the first place. okay. thanks Huang Shijie