From owner-freebsd-current@FreeBSD.ORG Fri Jul 10 05:45:41 2009 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id F335C106566B for ; Fri, 10 Jul 2009 05:45:41 +0000 (UTC) (envelope-from avg@freebsd.org) Received: from citadel.icyb.net.ua (citadel.icyb.net.ua [212.40.38.140]) by mx1.freebsd.org (Postfix) with ESMTP id 708EC8FC0C for ; Fri, 10 Jul 2009 05:45:38 +0000 (UTC) (envelope-from avg@freebsd.org) Received: from porto.topspin.kiev.ua (porto-e.starpoint.kiev.ua [212.40.38.100]) by citadel.icyb.net.ua (8.8.8p3/ICyb-2.3exp) with ESMTP id IAA25444; Fri, 10 Jul 2009 08:45:35 +0300 (EEST) (envelope-from avg@freebsd.org) Received: from localhost.topspin.kiev.ua ([127.0.0.1] helo=edge.pp.kiev.ua) by porto.topspin.kiev.ua with esmtp (Exim 4.34 (FreeBSD)) id 1MP8vO-000Kiy-Nz; Fri, 10 Jul 2009 08:45:34 +0300 Message-ID: <4A56D57D.1090508@freebsd.org> Date: Fri, 10 Jul 2009 08:45:33 +0300 From: Andriy Gapon User-Agent: Thunderbird 2.0.0.21 (X11/20090406) MIME-Version: 1.0 To: David Schultz , freebsd-current@freebsd.org References: <4A562960.3010801@freebsd.org> <20090709193932.GA54408@zim.MIT.EDU> In-Reply-To: <20090709193932.GA54408@zim.MIT.EDU> X-Enigmail-Version: 0.95.7 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Subject: Re: dtrace users opinion solicited (timestamps) X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 10 Jul 2009 05:45:42 -0000 on 09/07/2009 22:39 David Schultz said the following: > Doesn't Solaris DTRT and compensate for TSC frequency changes? > Why can't we do the same thing? That very well may be. I haven't thoroughly checked but I think that even we are doing the right thing when we use TSC as a timecounter. But at this moment I am looking for a quick/simple and "sufficiently good" fix for the bigger problems with DTrace timestamping (kern/127441). Another side of the "simplicity" is that the timestamp function for DTrace needs to be fast and should not itself be probed. -- Andriy Gapon