From owner-freebsd-arch Mon Feb 10 12:45:32 2003 Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 30C7937B401 for ; Mon, 10 Feb 2003 12:45:31 -0800 (PST) Received: from brev.stud.ntnu.no (brev.stud.ntnu.no [129.241.56.70]) by mx1.FreeBSD.org (Postfix) with ESMTP id ED82943FBD for ; Mon, 10 Feb 2003 12:45:27 -0800 (PST) (envelope-from morten@rodal.no) Received: from localhost (localhost [127.0.0.1]) by brev.stud.ntnu.no (Postfix) with ESMTP id 90B9713EE35; Mon, 10 Feb 2003 21:45:26 +0100 (CET) Received: from slurp.rodal.no (m200h.studby.ntnu.no [129.241.135.200]) by brev.stud.ntnu.no (Postfix) with ESMTP id A9D5713EE27; Mon, 10 Feb 2003 21:45:25 +0100 (CET) Received: (from morten@localhost) by slurp.rodal.no (8.12.6/8.12.6/Submit) id h1AKjOCt018340; Mon, 10 Feb 2003 21:45:24 +0100 (CET) (envelope-from morten) Date: Mon, 10 Feb 2003 21:45:24 +0100 From: Morten Rodal To: phk@phk.freebsd.dk Cc: David Schultz , arch@FreeBSD.ORG Subject: Re: Our lemming-syncer caught in the act. Message-ID: <20030210204523.GF12240@slurp.rodal.no> References: <20030210091317.GD5165@HAL9000.homeunix.com> <37473.1044868995@critter.freebsd.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <37473.1044868995@critter.freebsd.dk> X-Virus-Scanned: by AMaViS perl-11 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Mon, Feb 10, 2003 at 10:23:15AM +0100, phk@phk.freebsd.dk wrote: > In message <20030210091317.GD5165@HAL9000.homeunix.com>, David Schultz writes: > > >When a large file times out, a significant amount of I/O can be > >generated. This is still far better than the old syncer that > >flushed everything every 30 seconds. The reasons for this > >behavior are explained in src/sys/ufs/ffs/README. After reading > >that, do you still think it makes sense to try to do better? > > Yes, it makes a lot of sense. There is no point in batching up > writes to the point of showing 200 requests off at once then > wait 30 seconds, then do it again etc etc. > > We can and need to do better than that. > I am very sorry to pitch in on your discussion here, but I feel that this needs a little light. It should probably have been posted to -current as an own thread, but I felt that since you guys where already discussing the syncer I'd tell my tale. I upgraded my workstation to 5.0-RELEASE shortly after it became available, and there is one thing that keeps bugging me. When I delete a file that is large, say >600MB, the system performance and interactivity will drop remarkebly. I mean, after a little while (probably when the/syncer kicks in), the mouse starts to lag in X. I can type several letters before they appear, and xmms starts to skip while playing audio. Using top with the -S flag I can see syncer eating a lot of CPU and as soon as the syncer is done doing whatever it does everything goes back to normal. After I became aware of this syncer thingy I found out that whenever I hear skips in the music it is because the syncer uses CPU (according to top). Of couse this could be that I'm just using a shitty computer (it's an old dual p2-300mhz), but I never experienced this with 4.X series. As for the installation it's pretty standard. Softupdates on all mount points except root, no sysctl overrides or anything of that kind. This is, at least yet, the only thing that is really bugging me about 5.0. A fix for this would be really appreciated. -- Morten Rodal To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message