From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756276Ab3LEAyA (ORCPT ); Wed, 4 Dec 2013 19:54:00 -0500 Received: from mail-yh0-f49.google.com ([209.85.213.49]:49523 "EHLO mail-yh0-f49.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756105Ab3LEAx6 (ORCPT ); Wed, 4 Dec 2013 19:53:58 -0500 Date: Wed, 4 Dec 2013 16:53:54 -0800 (PST) From: David Rientjes X-X-Sender: rientjes@chino.kir.corp.google.com To: Christoph Lameter cc: Andrew Morton , Michal Hocko , KAMEZAWA Hiroyuki , Johannes Weiner , Mel Gorman , Rik van Riel , Pekka Enberg , linux-kernel@vger.kernel.org, linux-mm@kvack.org, cgroups@vger.kernel.org Subject: Re: [patch 3/8] mm, mempolicy: remove per-process flag In-Reply-To: <00000142be3633ba-2a459537-58fb-444b-a99f-33ff5e5b2aed-000000@email.amazonses.com> Message-ID: References: <20131119131400.GC20655@dhcp22.suse.cz> <20131119134007.GD20655@dhcp22.suse.cz> <20131120152251.GA18809@dhcp22.suse.cz> <20131128115458.GK2761@dhcp22.suse.cz> <00000142be3633ba-2a459537-58fb-444b-a99f-33ff5e5b2aed-000000@email.amazonses.com> User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) 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 On Wed, 4 Dec 2013, Christoph Lameter wrote: > > PF_MEMPOLICY is an unnecessary optimization for CONFIG_SLAB users. > > There's no significant performance degradation to checking > > current->mempolicy rather than current->flags & PF_MEMPOLICY in the > > allocation path, especially since this is considered unlikely(). > > The use of current->mempolicy increase the cache footprint since its in a > rarely used cacheline. This performance issue would occur when memory > policies are not used since that cacheline would then have to be touched > regardless of memory policies be in effect or not. PF_MEMPOLICY was used > to avoid touching the cacheline. > Right, but it turns out not to matter in practice. As one of the non- default CONFIG_SLAB users, and PF_MEMPOLICY only does something for CONFIG_SLAB, this patch tested to not show any degradation for specjbb which stresses the allocator in terms of throughput: with patch: 128761.54 SPECjbb2005 bops without patch: 127576.65 SPECjbb2005 bops These per-process flags are a scarce resource so I don't think PF_MEMPOLICY warrants a bit when it's not shown to be advantageous in configurations without mempolicy usage where it's intended to optimize, especially for a non-default slab allocator.