From mboxrd@z Thu Jan 1 00:00:00 1970 From: Kristoffer Glembo Subject: phylib printouts Date: Wed, 28 Oct 2009 17:33:19 +0100 Message-ID: <4AE8724F.3010104@gaisler.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit To: netdev@vger.kernel.org Return-path: Received: from mail176c2.megamailservers.com ([69.49.111.76]:42643 "EHLO mail176c2.megamailservers.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755138AbZJ1Qfz (ORCPT ); Wed, 28 Oct 2009 12:35:55 -0400 Received: from [192.168.0.35] (c-10a5e155.260-1-64736c10.cust.bredbandsbolaget.se [85.225.165.16]) (authenticated bits=0) by mail176c2.megamailservers.com (8.13.6/8.13.1) with ESMTP id n9SGZuIL005966 for ; Wed, 28 Oct 2009 12:35:58 -0400 Sender: netdev-owner@vger.kernel.org List-ID: Hi, There are some printouts in the phy abstraction library using pr_info(). Personally I would prefer if they used pr_debug() instead. Especially the pr_info("%s: probed\n", bus->name) in mdiobus_register() since in some cases I can't tolerate a console printout while the link is down (we have an ethernet debug link and read the UART output through the link). Of course this is a very specific case but in general in embedded systems (which I think is where the phylib is mainly used) I prefer the driver and other kernel code to be rather silent if not debugging. Any opinions? :) Best regards, Kristoffer Glembo