From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754218Ab1BUTJv (ORCPT ); Mon, 21 Feb 2011 14:09:51 -0500 Received: from mga09.intel.com ([134.134.136.24]:8560 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752082Ab1BUTIM (ORCPT ); Mon, 21 Feb 2011 14:08:12 -0500 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.62,201,1297065600"; d="scan'208";a="711415552" From: Andi Kleen To: Jack Steiner Cc: linux-mm@kvack.org, akpm@linux-foundation.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] - Improve drain pages performance on large systems References: <20110215223840.GA27420@sgi.com> Date: Tue, 15 Feb 2011 16:17:44 -0800 In-Reply-To: <20110215223840.GA27420@sgi.com> (Jack Steiner's message of "Tue, 15 Feb 2011 16:38:40 -0600") Message-ID: User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.2 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Jack Steiner writes: > Heavy swapping within a cpuset causes frequent calls to drain_all_pages(). I suspect drain_all_pages should be really made more zone aware in the first place and only drain what is actually needed (e.g. work off a zonelist). I was fighting with this for hwpoison too. That said your patch looks reasonable. -Andi -- ak@linux.intel.com -- Speaking for myself only