From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934078AbXCTVcZ (ORCPT ); Tue, 20 Mar 2007 17:32:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S934080AbXCTVcZ (ORCPT ); Tue, 20 Mar 2007 17:32:25 -0400 Received: from one.firstfloor.org ([213.235.205.2]:50834 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934078AbXCTVcY (ORCPT ); Tue, 20 Mar 2007 17:32:24 -0400 Date: Tue, 20 Mar 2007 22:32:18 +0100 From: Andi Kleen To: Christoph Lameter Cc: Eric Dumazet , Andrew Morton , Andi Kleen , linux kernel Subject: Re: [RFC] SLAB : NUMA cache_free_alien() very expensive because of virt_to_slab(objp); nodeid = slabp->nodeid; Message-ID: <20070320213218.GA13952@one.firstfloor.org> References: <20070320181235.77d28864.dada1@cosmosbay.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > > Is it possible virt_to_slab(objp)->nodeid being different from pfn_to_nid(objp) ? > > It is possible the page allocator falls back to another node than > requested. We would need to check that this never occurs. The only way to ensure that would be to set a strict mempolicy. But I'm not sure that's a good idea -- after all you don't want to fail an allocation in this case. But pfn_to_nid on the object like proposed by Eric should work anyways. But I'm not sure the tables used for that will be more often cache hot than the slab. -Andi