From owner-freebsd-questions@FreeBSD.ORG Mon Jun 16 11:35:28 2003 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4DE3637B4A3 for ; Mon, 16 Jun 2003 11:35:28 -0700 (PDT) Received: from rosebud.alerce.com (rosebud.lbl.gov [131.243.193.115]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5BC7F43F93 for ; Mon, 16 Jun 2003 11:35:27 -0700 (PDT) (envelope-from hartzell@rosebud.alerce.com) Received: from rosebud.alerce.com (localhost [127.0.0.1]) by rosebud.alerce.com (8.12.8p1/8.12.8) with ESMTP id h5GIZQSv014090 for ; Mon, 16 Jun 2003 11:35:26 -0700 (PDT) (envelope-from hartzell@rosebud.alerce.com) Received: (from hartzell@localhost) by rosebud.alerce.com (8.12.8p1/8.12.8/Submit) id h5GIZPWl014058; Mon, 16 Jun 2003 11:35:25 -0700 (PDT) From: George Hartzell MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <16110.3565.71625.373384@rosebud.alerce.com> Date: Mon, 16 Jun 2003 11:35:25 -0700 To: freebsd-questions@freebsd.org X-Mailer: VM 7.14 under 21.4 (patch 12) "Portable Code" XEmacs Lucid Subject: Using kevent to catch laptop disk I/O culprit? X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: George Hartzell List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 16 Jun 2003 18:35:28 -0000 I'm working on keeping my laptop disk spun down and am using the process of elimination to figure out who's always spinning up the disk. So far, I've put /tmp on an mfs, and I just installed Robert Sexton's hw.ata.suspend patch from the freebsd-mobile archives. I'm slowly killing off and/or fine tuning various daemons to see if they're responsible for all of the IO, and it occured to me that the kevent/kqueue stuff might be a great way to actually see who's writing what, when. Has anyone written the necessary parts to get this kind of a simple filesystem trace? g.