Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932322AbZD1Rp2 (ORCPT ); Tue, 28 Apr 2009 13:45:28 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1757660AbZD1RpJ (ORCPT ); Tue, 28 Apr 2009 13:45:09 -0400 Received: from waste.org ([66.93.16.53]:58683 "EHLO waste.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755540AbZD1RpH (ORCPT ); Tue, 28 Apr 2009 13:45:07 -0400 Subject: Re: [PATCH 5/5] proc: export more page flags in /proc/kpageflags From: Matt Mackall To: Ingo Molnar Cc: Pekka Enberg , Andi Kleen , Wu Fengguang , Steven Rostedt , =?ISO-8859-1?Q?Fr=E9d=E9ric?= Weisbecker , Larry Woodman , Peter Zijlstra , Eduard - Gabriel Munteanu , Andrew Morton , LKML , KOSAKI Motohiro , Alexey Dobriyan , "linux-mm@kvack.org" In-Reply-To: <20090428093621.GD21085@elte.hu> References: <20090428010907.912554629@intel.com> <20090428014920.769723618@intel.com> <20090428065507.GA2024@elte.hu> <20090428074031.GK27382@one.firstfloor.org> <1240909484.1982.16.camel@penberg-laptop> <20090428091508.GA21085@elte.hu> <84144f020904280219p197d5ceag846ae9a80a76884e@mail.gmail.com> <84144f020904280225h490ef682p8973cb1241a1f3ea@mail.gmail.com> <20090428093621.GD21085@elte.hu> Content-Type: text/plain Date: Tue, 28 Apr 2009 12:42:08 -0500 Message-Id: <1240940528.938.426.camel@calx> Mime-Version: 1.0 X-Mailer: Evolution 2.24.5 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1289 Lines: 39 On Tue, 2009-04-28 at 11:36 +0200, Ingo Molnar wrote: > I 'integrate' traces all the time to get summary counts. This series > of dynamic events: > > allocation > page count up > page count up > page count down > page count up > page count up > page count up > page count up > > integrates into: "page count is 6". Perhaps you've failed calculus. The integral is 6 + C. This is a critical distinction. Tracing is great for looking at changes, but it completely falls down for static system-wide measurements because it would require integrating from time=0 to get a meaningful summation. That's completely useless for taking a measurement on a system that already has an uptime of months. Never mind that summing up page flag changes for every page on the system since boot time through the trace interface is incredibly wasteful given that we're keeping a per-page integral in the page tables anyway. Tracing is not the answer for everything. -- http://selenic.com : development and support for Mercurial and Linux -- 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/