From owner-freebsd-current@FreeBSD.ORG Fri Jun 10 02:30:55 2005 Return-Path: X-Original-To: freebsd-current@FreeBSD.org Delivered-To: freebsd-current@FreeBSD.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5616216A41C for ; Fri, 10 Jun 2005 02:30:55 +0000 (GMT) (envelope-from truckman@FreeBSD.org) Received: from gw.catspoiler.org (217-ip-163.nccn.net [209.79.217.163]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1892543D48 for ; Fri, 10 Jun 2005 02:30:54 +0000 (GMT) (envelope-from truckman@FreeBSD.org) Received: from FreeBSD.org (mousie.catspoiler.org [192.168.101.2]) by gw.catspoiler.org (8.13.1/8.13.1) with ESMTP id j5A2UaBx028568; Thu, 9 Jun 2005 19:30:41 -0700 (PDT) (envelope-from truckman@FreeBSD.org) Message-Id: <200506100230.j5A2UaBx028568@gw.catspoiler.org> Date: Thu, 9 Jun 2005 19:30:36 -0700 (PDT) From: Don Lewis To: emil@cs.rmit.edu.au In-Reply-To: <20050610020331.GA1264@cs.rmit.edu.au> MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Cc: freebsd-current@FreeBSD.org Subject: Re: Sound skipping problems 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 Jun 2005 02:30:55 -0000 On 10 Jun, Emil Mikulic wrote: > On Thu, Jun 09, 2005 at 06:06:28PM +0900, Pyun YongHyeon wrote: >> Wow! It helps a lot. > > Yeah, it's awesome! Great work, Jeff! > >> Even if maestro3(4) does not support channel_setblocksize I couldn't >> notice audio stuttering while extracting mozilla source. I noticed a >> few sound skipping when a command "rm -rf mozilla" was in progress. > > I notice the occasional stutter towards the end of "make clean" in > ports/www/firefox. Unless I'm reading it wrong, schedgraph is showing > that this is because of syncer, not bufdaemon. Not too suprising. The syncer probably call VOP_FSYNC() on one of the syncer vnodes, which walks the potentially very lengthy per-mount vnode list in order to handle inode timestamp updates. It would be a lot more efficient to handle the timestamp updates by putting the vnodes on the syncer work list and it would avoid the periodic massive bursts of file system activity. I started working on this quite a while ago, but ran into a few problems and ran out of time to work on it.