From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755022Ab1LMTGn (ORCPT ); Tue, 13 Dec 2011 14:06:43 -0500 Received: from mga11.intel.com ([192.55.52.93]:60045 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751821Ab1LMTGl (ORCPT ); Tue, 13 Dec 2011 14:06:41 -0500 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.71,315,1320652800"; d="scan'208";a="95752254" Date: Tue, 13 Dec 2011 11:06:32 -0800 From: Andi Kleen To: Shaohua Li Cc: lkml , linux-mm , Andrew Morton , Jens Axboe , Christoph Lameter , lee.schermerhorn@hp.com, David Rientjes Subject: Re: [patch v3]numa: add a sysctl to control interleave allocation granularity from each node to improve I/O performance Message-ID: <20111213190632.GA5830@tassilo.jf.intel.com> References: <1323655125.22361.376.camel@sli10-conroe> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1323655125.22361.376.camel@sli10-conroe> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Dec 12, 2011 at 09:58:45AM +0800, Shaohua Li wrote: > If mem plicy is interleaves, we will allocated pages from nodes in a round > robin way. This surely can do interleave fairly, but not optimal. > > Say the pages will be used for I/O later. Interleave allocation for two pages > are allocated from two nodes, so the pages are not physically continuous. Later I would prefer to add a new policy (INTERLEAVE_MULTI or so) for this instead of a global sysctl, that takes the additional parameter. Also I don't like having more per task state. Could you compute this from the address instead even for the process policy case? -Andi