From owner-freebsd-stable@FreeBSD.ORG Thu Aug 19 01:49:44 2004 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5258416A4CE for ; Thu, 19 Aug 2004 01:49:44 +0000 (GMT) Received: from carver.gumbysoft.com (carver.gumbysoft.com [66.220.23.50]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4047E43D31 for ; Thu, 19 Aug 2004 01:49:44 +0000 (GMT) (envelope-from dwhite@gumbysoft.com) Received: by carver.gumbysoft.com (Postfix, from userid 1000) id 301FE72DD4; Wed, 18 Aug 2004 18:49:44 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by carver.gumbysoft.com (Postfix) with ESMTP id 2AE6272DCB; Wed, 18 Aug 2004 18:49:44 -0700 (PDT) Date: Wed, 18 Aug 2004 18:49:44 -0700 (PDT) From: Doug White To: Brian Astill In-Reply-To: <200408180845.45217.bastill@adam.com.au> Message-ID: <20040818184452.E55263@carver.gumbysoft.com> References: <20040817104030.A40914@carver.gumbysoft.com> <200408180845.45217.bastill@adam.com.au> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: Vasily cc: freebsd-stable@freebsd.org Subject: Re: microuptime() ... X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 19 Aug 2004 01:49:44 -0000 On Wed, 18 Aug 2004, Brian Astill wrote: > > Because VIA doesn't publish documentaion, and source is not > > documentation. > > I don't quite understand this response. Are you saying that the Linux > kernel team are somehow privileged cpw FBSD, so that they can avoid > that microuptime issue and FBSD can't? I would have thought that the > source cpw documentation for VIA chipsets would have been the same for > all. What's 'cpw'? Criminal Posession of a Weapon? :) I can't speak to Linux's timecounter implementation since I haven't read their code. I suspect they don't have any sort of thing like microuptime(), using jiffies insteaed which are somewhat lower precision. They may just ignore the any timecounter disruptions due to CPU clock adjustments, for all I know. And it is *very* possible someone inside of VIA or one of their customers got access to the docs that we can't get to and added the requisite code. If you can derive the appropriate bits and offer them as a patch then all the better. Linux has lots of friends in high places since its Speshul. -- Doug White | FreeBSD: The Power to Serve dwhite@gumbysoft.com | www.FreeBSD.org