From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1161107AbXDJXsd (ORCPT ); Tue, 10 Apr 2007 19:48:33 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1161104AbXDJXsc (ORCPT ); Tue, 10 Apr 2007 19:48:32 -0400 Received: from mga02.intel.com ([134.134.136.20]:27917 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1161107AbXDJXsc (ORCPT ); Tue, 10 Apr 2007 19:48:32 -0400 X-ExtLoop1: 1 X-IronPort-AV: i="4.14,392,1170662400"; d="scan'208"; a="225249669:sNHT3355638083" Date: Tue, 10 Apr 2007 16:47:06 -0700 From: "Siddha, Suresh B" To: Ingo Molnar Cc: Andrew Morton , "Siddha, Suresh B" , nickpiggin@yahoo.com.au, linux-kernel@vger.kernel.org, Andi Kleen , Ravikiran G Thirumalai Subject: Re: [patch] sched: align rq to cacheline boundary Message-ID: <20070410234706.GA585@linux-os.sc.intel.com> References: <20070409180853.GC3948@linux-os.sc.intel.com> <20070409134057.2d249f0c.akpm@linux-foundation.org> <20070410063656.GB23274@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070410063656.GB23274@elte.hu> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Apr 10, 2007 at 08:36:56AM +0200, Ingo Molnar wrote: > the runqueue is really supposed to be cacheline-isolated at _both_ ends > - at its beginning and at its end as well. Then either we need to define the first element in the struct as cacheline aligned or move into section where all the elements are cacheline aligned. My current patch will not align it at the end. > > Remember also that the linesize on VSMP is 4k. > > that sucks ... > > maybe, to mitigate some of the costs, do a special PER_CPU_CACHE_ALIGNED > area that collects per-cpu fields that also have significant cross-CPU > use and need cacheline isolation? Such cacheline-aligned variables, if > collected separately, would pack up more tightly and would cause only > half of the wasted space. I will post a patch(once Jermey's page align percpu patch goes in to -mm) thanks, suresh