From owner-freebsd-stable@FreeBSD.ORG Mon Oct 31 23:54:38 2005 Return-Path: X-Original-To: stable@freebsd.org 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 5048116A41F for ; Mon, 31 Oct 2005 23:54:38 +0000 (GMT) (envelope-from rwatson@FreeBSD.org) Received: from cyrus.watson.org (cyrus.watson.org [209.31.154.42]) by mx1.FreeBSD.org (Postfix) with ESMTP id EFF2343DBA for ; Mon, 31 Oct 2005 23:54:37 +0000 (GMT) (envelope-from rwatson@FreeBSD.org) Received: from fledge.watson.org (fledge.watson.org [209.31.154.41]) by cyrus.watson.org (Postfix) with ESMTP id 0F34A46BA4; Mon, 31 Oct 2005 18:54:37 -0500 (EST) Date: Mon, 31 Oct 2005 23:54:36 +0000 (GMT) From: Robert Watson X-X-Sender: robert@fledge.watson.org To: Marc Olzheim In-Reply-To: <20051031151809.GA83253@stack.nl> Message-ID: <20051031235311.D18382@fledge.watson.org> References: <436116D3.908@samsco.org> <20051031151809.GA83253@stack.nl> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: stable@freebsd.org Subject: Re: HEADS UP! 6.0-RELEASE coming X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 31 Oct 2005 23:54:38 -0000 On Mon, 31 Oct 2005, Marc Olzheim wrote: > On Thu, Oct 27, 2005 at 12:05:07PM -0600, Scott Long wrote: >> Everyone is still welcome to update their sources on the RELENG_6_0 >> branch and provide feedback for the next 48 hours or so. The release >> will likely be announced by the end of the weekend or early next >> week, at the latest. > > Anyone having updates for my showstopper page, please, notify me... > > http://www.stack.nl/%7Emarcolz/FreeBSD/showstoppers.html > > We are still installing new servers with 4.x, because I haven't seen a > really stable FreeBSD-stable release since... This includes current > RELENG_6_0. I started going through the ktrace code a few days ago to look at adding thread IDs to ktrace records. I'm happy to pick up the ktrace item in your todo list. My plan is basically to restore elements of the 4.x queueing model: rather than asynchronously dispatching all ktrace records to a single worker thread, dispatch them to disk from the system call return path for a thread. There are a few issues with threading that require attention -- in particular, avoiding writes on the vnode operation after close. This will also slow down ktracing again, but should avoid dropping records. I hope to have something hacked together in a few days, but not in time for 6.0. Robert N M Watson