From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753517AbbCWWTZ (ORCPT ); Mon, 23 Mar 2015 18:19:25 -0400 Received: from e31.co.us.ibm.com ([32.97.110.149]:50722 "EHLO e31.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752748AbbCWWTW (ORCPT ); Mon, 23 Mar 2015 18:19:22 -0400 Date: Mon, 23 Mar 2015 15:17:20 -0700 From: Sukadev Bhattiprolu To: Michael Ellerman Cc: Paul Mackerras , peterz@infradead.org, linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org Subject: Re: [2/9] powerpc/hv24x7: Remove unnecessary parameter Message-ID: <20150323221720.GB6013@us.ibm.com> References: <1424210434-28070-3-git-send-email-sukadev@linux.vnet.ibm.com> <20150317001352.69BC914010F@ozlabs.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150317001352.69BC914010F@ozlabs.org> X-Operating-System: Linux 2.0.32 on an i486 User-Agent: Mutt/1.5.21 (2010-09-15) X-TM-AS-MML: disable X-Content-Scanned: Fidelis XPS MAILER x-cbid: 15032322-8236-0000-0000-00000A53ED11 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Michael Ellerman [mpe@ellerman.id.au] wrote: | On Tue, 2015-17-02 at 22:00:27 UTC, Sukadev Bhattiprolu wrote: | > Use pr_notice_ratelimited() to log error messages and remove | > the 'success_expected' parameter. | | I don't understand how this is equivalent? They are two unrelated changes that I should have separated. | | The current code uses success_expected to indicate that once it's done the | request once and found that it works, it then expects the request to continue | working, and if it doesn't then that is an error. The current code is using success_expected to _not_ log an error if that initial request fails. i.e we silently return -EIO here. I think the 'success_expected' parameter is not really necessary. We can simply log the message even for that initial request. And we can log it a lower priority than KERN_ERR since the message is mostly for developers rather than users who would use event names (which encode/abstract the domain and offset values). | | Using pr_ratelimited() will do the opposite, ie. the first failure will print a | message, but that may not really indicate an error, it may just be a badly | configured request. | | Or at least that's how I understand it, please convince me I'm wrong :) | | cheers | _______________________________________________ | Linuxppc-dev mailing list | Linuxppc-dev@lists.ozlabs.org | https://lists.ozlabs.org/listinfo/linuxppc-dev