Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261263AbUDGXXY (ORCPT ); Wed, 7 Apr 2004 19:23:24 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261234AbUDGXXY (ORCPT ); Wed, 7 Apr 2004 19:23:24 -0400 Received: from e6.ny.us.ibm.com ([32.97.182.106]:22659 "EHLO e6.ny.us.ibm.com") by vger.kernel.org with ESMTP id S261263AbUDGXXW (ORCPT ); Wed, 7 Apr 2004 19:23:22 -0400 Date: Wed, 07 Apr 2004 16:34:51 -0700 From: "Martin J. Bligh" To: Andrea Arcangeli cc: Ingo Molnar , Eric Whiting , akpm@osdl.org, linux-kernel@vger.kernel.org Subject: Re: -mmX 4G patches feedback [numbers: how much performance impact] Message-ID: <33900000.1081380891@flay> In-Reply-To: <20040407231806.GV26888@dualathlon.random> References: <40718B2A.967D9467@amis.com> <20040405174616.GH2234@dualathlon.random> <4071D11B.1FEFD20A@amis.com> <20040405221641.GN2234@dualathlon.random> <20040406115539.GA31465@elte.hu> <20040406155925.GW2234@dualathlon.random> <20040406192549.GA14869@elte.hu> <12640000.1081378705@flay> <20040407230140.GT26888@dualathlon.random> <29510000.1081380104@flay> <20040407231806.GV26888@dualathlon.random> X-Mailer: Mulberry/2.1.2 (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 Content-Length: 1166 Lines: 24 --On Thursday, April 08, 2004 01:18:06 +0200 Andrea Arcangeli wrote: > On Wed, Apr 07, 2004 at 04:21:44PM -0700, Martin J. Bligh wrote: >> Speaking of which, pte_highmem is stinking expensive itself. There's >> probably a large class of workloads that'd work with out pte_highmem >> if we had 4/4 split (or shared pagetables. Grrr ;-)) > > hey, I can add a sysctl in 5 minutes to disable pte_highmem at runtime, > why do you think it's expensive, it should be not, it's all atomic kmaps > only doing invlpg. The few workloads trashing on the ptes manipulation > needs pte_highmem anyways. If I thought it was expensive for any common > load the sysctl would be already there. I measured it - IIRC it was 5-10% on kernel compile ... and that was on a high ratio NUMA which it should have made *better* (as with highmem, the PTEs can be allocated node locally). I'll try to dig up the old profiles. M. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/