Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932118Ab2ERQEU (ORCPT ); Fri, 18 May 2012 12:04:20 -0400 Received: from merlin.infradead.org ([205.233.59.134]:36482 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757788Ab2ERQET convert rfc822-to-8bit (ORCPT ); Fri, 18 May 2012 12:04:19 -0400 Message-ID: <1337357041.573.86.camel@twins> Subject: Re: [tip:sched/numa] sched/numa: Introduce sys_numa_{t,m}bind() From: Peter Zijlstra To: Christoph Lameter Cc: Rik van Riel , mingo@kernel.org, hpa@zytor.com, linux-kernel@vger.kernel.org, torvalds@linux-foundation.org, pjt@google.com, bharata.rao@gmail.com, akpm@linux-foundation.org, Lee.Schermerhorn@hp.com, aarcange@redhat.com, danms@us.ibm.com, suresh.b.siddha@intel.com, tglx@linutronix.de, linux-tip-commits@vger.kernel.org Date: Fri, 18 May 2012 18:04:01 +0200 In-Reply-To: References: <4FB66756.2060302@redhat.com> <1337355341.573.68.camel@twins> <1337355610.573.70.camel@twins> <1337356192.573.75.camel@twins> Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT X-Mailer: Evolution 3.2.2- Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1089 Lines: 30 On Fri, 2012-05-18 at 11:00 -0500, Christoph Lameter wrote: > Having the page local is a win if there are a sufficient number of > accesses to amortize the effort to move the page. Given the expensive > nature of page migration there would need to be a large number of accesses > to a page to justify the effort. Right, but this is true of any migration scheme and not specific to MoF. > > How does it matter how you migrate? > > Migrate on fault incurs two types of costs: > > 1. Unmapping. This results in additional faults to reestablish the ptes. > > 2. Actual lazy migrate. More faults. Now the page needs to be copied to > the new node and the actual migration work is done. Nah, only the 1 fault is extra. Regular migration already needs to unmap and copy and reinstate, so the only extra work is the fault to trigger it. -- 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/