From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753385Ab1GOQIh (ORCPT ); Fri, 15 Jul 2011 12:08:37 -0400 Received: from mail.perches.com ([173.55.12.10]:4023 "EHLO mail.perches.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750775Ab1GOQIg (ORCPT ); Fri, 15 Jul 2011 12:08:36 -0400 Subject: Re: [PATCH 11/11] dynamic_debug: use a single printk() to emit msgs From: Joe Perches To: Bart Van Assche Cc: Jason Baron , gregkh@suse.de, jim.cromie@gmail.com, linux-kernel@vger.kernel.org In-Reply-To: References: <3667c0a87dd8fd64fdf1b1e8107b130a9b41096e.1310657068.git.jbaron@redhat.com> <1310744914.7582.36.camel@Joe-Laptop> Content-Type: text/plain; charset="UTF-8" Date: Fri, 15 Jul 2011 09:08:34 -0700 Message-ID: <1310746114.7582.45.camel@Joe-Laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.32.2 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 2011-07-15 at 17:57 +0200, Bart Van Assche wrote: > On Fri, Jul 15, 2011 at 5:48 PM, Joe Perches wrote: > > On Fri, 2011-07-15 at 12:05 +0200, Bart Van Assche wrote: > >> Not that's it important, but this change makes is possible to > >> eliminate the tid[] and lineno[] arrays again. Has that been > >> considered ? > > tid and lineno are decimal. Not using intermediate > > arrays would require awkward contortions to snprintf > > them without emitting 0 in the output. > Sorry, but I do not agree with the above. Hey Bart. Then I think you don't understand the goal of the change. The goal is to avoid possible interleaving of multiple printks from different threads. > The current implementation > of __dynamic_pr_debug() shows that an implementation without temporary > arrays does not require any awkward constructs. See e.g. > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=lib/dynamic_debug.c > for the source code of the current implementation of > __dynamic_pr_debug(). > This may be subjective, but my opinion is that following that style > results in shorter and more elegant code than the approach with the > temporary arrays proposed in patch 11/11. It requires awkward contortions to emit a decimal value or "" without multiple printks or temporaries. cheers, Joe