From owner-freebsd-current Mon Jan 25 18:07:48 1999 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id SAA28287 for freebsd-current-outgoing; Mon, 25 Jan 1999 18:07:48 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from janus.syracuse.net (janus.syracuse.net [205.232.47.15]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA28233 for ; Mon, 25 Jan 1999 18:07:39 -0800 (PST) (envelope-from green@unixhelp.org) Received: from localhost (green@localhost) by janus.syracuse.net (8.8.8/8.8.7) with ESMTP id VAA12512; Mon, 25 Jan 1999 21:07:16 -0500 (EST) Date: Mon, 25 Jan 1999 21:07:15 -0500 (EST) From: Brian Feldman X-Sender: green@janus.syracuse.net To: Matthew Dillon cc: Sheldon Hearn , Lee Cremeans , current@FreeBSD.ORG Subject: Re: IDE DMA works, I'll be a... In-Reply-To: <199901260135.RAA11762@apollo.backplane.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Mon, 25 Jan 1999, Matthew Dillon wrote: > :> I would like to know if your transfer rate improves or not, and by > :> how much. > :> > :> #if 0 > :> if (cnt.v_inactive_count / 3 > page_shortage) { > :> maxlaunder = 0; > :> launder_loop = 0; > :> } else > :> #endif > :> { > :> -Matt > : > :I'd rather use if (0 && cnt.v_inactive_count / 3 > page_shortage) {. I'm > :trying it out now, here's the "before": > : > :IOZONE performance measurements: > : 2972048 bytes/second for writing the file > : 2962863 bytes/second for reading the file > : Brian Feldman _ __ ___ ___ ___ > > Welll... I went ahead and tested it on my diskless workstation, > which actually has a DMA IDE drive connected to it at the moment > for another test I'm running. > > I didn't see any change in performance. That doesn't meant that > hasn't been a chance, just that it doesn't look like the one thing > I thought might be causing it is causing it. > > Are you sure the problem isn't simply that your disk is getting > a bit more full and causing the test to skip around more ( or move > to more inner tracks, which have lower transfer rates), and > thus appear to slow down a little ? No, this REALLY isn't it. I have tons free on the test drive. iozone 100 on a 64 mb system: {"/home/green"}$ diff iozone.old iozone 15,16c15,16 < Writing the 100 Megabyte file, 'iozone.tmp'...37.320312 seconds < Reading the file...37.710938 seconds --- > Writing the 100 Megabyte file, 'iozone.tmp'...37.750000 seconds > Reading the file...37.687500 seconds 19,20c19,20 < 2809665 bytes/second for writing the file < 2780562 bytes/second for reading the file --- > 2777684 bytes/second for writing the file > 2782291 bytes/second for reading the file > > -Matt > > Before ( iozone 64 on a 48MB machine ) > > IOZONE performance measurements: > 7809031 bytes/second for writing the file > 10324440 bytes/second for reading the file > > IOZONE performance measurements: > 8388608 bytes/second for writing the file > 10324440 bytes/second for reading the file > > After ( iozone 64 on a 48MB machine ) > > IOZONE performance measurements: > 8103711 bytes/second for writing the file > 10336864 bytes/second for reading the file > > IOZONE performance measurements: > 8429768 bytes/second for writing the file > 10324440 bytes/second for reading the file > > Matthew Dillon > > Brian Feldman _ __ ___ ___ ___ green@unixhelp.org _ __ ___ | _ ) __| \ http://www.freebsd.org/ _ __ ___ ____ | _ \__ \ |) | FreeBSD: The Power to Serve! _ __ ___ ____ _____ |___/___/___/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message