From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262601AbTKIP6m (ORCPT ); Sun, 9 Nov 2003 10:58:42 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262608AbTKIP6m (ORCPT ); Sun, 9 Nov 2003 10:58:42 -0500 Received: from citrine.spiritone.com ([216.99.193.133]:31387 "EHLO citrine.spiritone.com") by vger.kernel.org with ESMTP id S262601AbTKIP6k (ORCPT ); Sun, 9 Nov 2003 10:58:40 -0500 Date: Sun, 09 Nov 2003 07:58:19 -0800 From: "Martin J. Bligh" To: Davide Libenzi cc: Nick Piggin , Andrew Morton , Con Kolivas , Linux Kernel Mailing List , Ingo Molnar Subject: Re: [PATCH] Fix find busiest queue 2.6.0-test9 Message-ID: <122840000.1068393498@[10.10.2.4]> In-Reply-To: References: X-Mailer: Mulberry/2.2.1 (Linux/x86) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org >> I ran it on the 16-way - no difference in performance. If the code is >> correct as was before (and I agree, it seems it was), perhaps it's just >> in need of a big fat comment to explain the confusion? ;-) > > Ingo already dropped a fat comment ;) This is the relevant part: > > * We fend off statistical fluctuations in runqueue lengths by > * saving the runqueue length during the previous load-balancing > * operation and using the smaller one the current and saved lengths. I think the confusing bit is this: this_rq->prev_cpu_load[i] = rq_src->nr_running; where "this_rq->prev_cpu_load[i]" doesn't intuitively look like what it means ;-) Even just 's/i/cpu/' would help a bit, or something (like wrapping it in macros). Seems it is correct as was though - thanks for explaining it. M.