From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S266274AbUHBGAZ (ORCPT ); Mon, 2 Aug 2004 02:00:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S266275AbUHBGAZ (ORCPT ); Mon, 2 Aug 2004 02:00:25 -0400 Received: from ausmtp02.au.ibm.com ([202.81.18.187]:57028 "EHLO ausmtp02.au.ibm.com") by vger.kernel.org with ESMTP id S266274AbUHBGAX (ORCPT ); Mon, 2 Aug 2004 02:00:23 -0400 Subject: Re: [PATCH] use for_each_cpu From: Rusty Russell To: Andrew Morton Cc: Anton Blanchard , lkml - Kernel Mailing List In-Reply-To: <20040801004708.6fa9f6f8.akpm@osdl.org> References: <20040801060144.GI30253@krispykreme> <20040731230859.138ba584.akpm@osdl.org> <20040801072711.GJ30253@krispykreme> <20040801004708.6fa9f6f8.akpm@osdl.org> Content-Type: text/plain Message-Id: <1091401715.24784.3.camel@bach> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.6 Date: Mon, 02 Aug 2004 15:59:55 +1000 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 2004-08-01 at 17:47, Andrew Morton wrote: > > Is it worth adding complexity to the cpu > > notifiers vs just using for_each_cpu? > > yup ;) It's only six lines, and it follows the same pattern as is used in, > say, page_alloc_cpu_notify(). Doing the same thing the same way in > multiple places is to be preferred, yes? Yes, and that way is "for_each_cpu". Andrew, take his patch. You know for_each_cpu is preferred over for_each_online_cpu unless there's a good reason for the latter. Rusty. -- Anyone who quotes me in their signature is an idiot -- Rusty Russell