From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934213Ab2GLRvw (ORCPT ); Thu, 12 Jul 2012 13:51:52 -0400 Received: from mx1.redhat.com ([209.132.183.28]:31169 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932781Ab2GLRvu (ORCPT ); Thu, 12 Jul 2012 13:51:50 -0400 Date: Thu, 12 Jul 2012 19:50:15 +0200 From: Andrea Arcangeli To: Konrad Rzeszutek Wilk Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, Hillf Danton , Dan Smith , Peter Zijlstra , Linus Torvalds , Andrew Morton , Thomas Gleixner , Ingo Molnar , Paul Turner , Suresh Siddha , Mike Galbraith , "Paul E. McKenney" , Lai Jiangshan , Bharata B Rao , Lee Schermerhorn , Rik van Riel , Johannes Weiner , Srivatsa Vaddagiri , Christoph Lameter , Alex Shi , Mauricio Faria de Oliveira , Konrad Rzeszutek Wilk , Don Morris , Benjamin Herrenschmidt Subject: Re: [PATCH 18/40] autonuma: call autonuma_setup_new_exec() Message-ID: <20120712175015.GJ20382@redhat.com> References: <1340888180-15355-1-git-send-email-aarcange@redhat.com> <1340888180-15355-19-git-send-email-aarcange@redhat.com> <20120630050425.GE3975@localhost.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120630050425.GE3975@localhost.localdomain> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi, On Sat, Jun 30, 2012 at 01:04:26AM -0400, Konrad Rzeszutek Wilk wrote: > On Thu, Jun 28, 2012 at 02:55:58PM +0200, Andrea Arcangeli wrote: > > This resets all per-thread and per-process statistics across exec > > syscalls or after kernel threads detached from the mm. The past > > statistical NUMA information is unlikely to be relevant for the future > > in these cases. > > The previous patch mentioned that it can run in bypass mode. Is > this also able to do so? Meaning that these calls end up doing nops? execve is always resetting unconditionally, there's no option to inherit the stats from the program running execve. By default all thread and mm stats are also resetted across fork/clone but there's a way to disable that with sysfs because those are more likely to retain some statistical significance.