From owner-freebsd-hardware Sun Apr 19 01:55:36 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA19648 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 01:55:36 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from vnode.vmunix.com (vnode.vmunix.com [209.112.4.20]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id IAA19643 for ; Sun, 19 Apr 1998 08:55:33 GMT (envelope-from mark@vnode.vmunix.com) Received: (from mark@localhost) by vnode.vmunix.com (8.8.8/8.8.8) id EAA03757; Sun, 19 Apr 1998 04:56:51 -0400 (EDT) (envelope-from mark) Message-ID: <19980419045651.63169@vmunix.com> Date: Sun, 19 Apr 1998 04:56:51 -0400 From: Mark Mayo To: Bruce Evans Cc: hardware@FreeBSD.ORG Subject: Re: best wdc0 flags ? References: <199804190652.QAA23558@godzilla.zeta.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89.1i In-Reply-To: <199804190652.QAA23558@godzilla.zeta.org.au>; from Bruce Evans on Sun, Apr 19, 1998 at 04:52:34PM +1000 X-Operating-System: FreeBSD 2.2.6-STABLE i386 Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Apr 19, 1998 at 04:52:34PM +1000, Bruce Evans wrote: > >Damn.. I just tried doing a > > dd if=/dev/rwd0 of=/dev/null bs=1m count=1280 > >Any ideas what's going wrong here? > > Your BIOS is misconfigured or only supports slow PIO modes. I use AUTO > mode and checked that this gave 16.6MB/sec by looking at `systat -vmstat' > (interrupt overhead should be about 10/16.6 * 100% for a 10 MB/sec drive). > My BIOS also supports PIO modes 0, 1, 2, 3 and 4. The 10MB firehose > swamps at least PIO modes 0 and 1. I see what's happening now.. The boot drive on this system is a SCSI disk, and when I added the IDE drive the BIOS absolutely refused to boot off the SCSI (even though it was set to..) -- so I set the "primary master" to "NONE" and let the BIOS think there was only the SCSI. Naturally, FreeBSD picked up the IDE just fine, but apparently the controller is operating in PIO mode 0 or 1, and as you say it just can't keep up with the Fireball... Ugly situation. :-) The machine is remote, so playing around more is a pain. Perhaps I'll try setting the BIOS to "auto" like you said, but not run the IDE AutoDetect thing and see if that works.. otherwise I don't know how I can get it to boot off the damned SCSI drive without using a floppy (which makes me nervous since the box is far enough away to be a real pain to get on the console..) Thanks for the superb explanation. -Mark > > Bruce -- ------------------------------------------------------------------------ Mark Mayo mark@vmunix.com RingZero Comp. http://www.vmunix.com/mark finger mark@vmunix.com for my PGP key and GCS code ------------------------------------------------------------------------ "The problem is how do you build tools that understand your programs at a deeper semantic level." - James Gosling To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 03:13:11 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id DAA29418 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 03:13:11 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from shell.monmouth.com (root@shell.monmouth.com [205.231.236.9]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id KAA29413 for ; Sun, 19 Apr 1998 10:13:08 GMT (envelope-from wilson@monmouth.com) Received: from monmouth.com (dn-max-ppp28.monmouth.com [205.229.220.105]) by shell.monmouth.com (8.8.5/8.7.3) with ESMTP id GAA04209 for ; Sun, 19 Apr 1998 06:12:38 -0400 (EDT) Message-ID: <3539CE31.6D8B5DF7@monmouth.com> Date: Sun, 19 Apr 1998 06:13:05 -0400 From: "W. Wilson" X-Mailer: Mozilla 4.05 [en] (X11; I; FreeBSD 2.2.5-RELEASE i386) MIME-Version: 1.0 To: freebsd-hardware@FreeBSD.ORG Subject: (no subject) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org who freebsd-hardware info To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 05:00:01 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id FAA11103 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 05:00:01 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from piggy.mdstud.chalmers.se (root@piggy.mdstud.chalmers.se [129.16.234.10]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA11062 for ; Sun, 19 Apr 1998 11:59:57 GMT (envelope-from md6tommy@mdstud.chalmers.se) Received: from grosse.mdstud.chalmers.se (md6tommy@grosse.mdstud.chalmers.se [129.16.234.21]) by piggy.mdstud.chalmers.se (8.8.5/8.8.5) with ESMTP id NAA06195; Sun, 19 Apr 1998 13:59:42 +0200 (MET DST) Received: from localhost (md6tommy@localhost) by grosse.mdstud.chalmers.se (8.8.5/8.8.5) with SMTP id NAA21969; Sun, 19 Apr 1998 13:59:40 +0200 (MET DST) X-Authentication-Warning: grosse.mdstud.chalmers.se: md6tommy owned process doing -bs Date: Sun, 19 Apr 1998 13:59:39 +0200 (MET DST) From: Tommy Hallgren To: hardware@FreeBSD.ORG cc: Mark Mayo Subject: Re: best wdc0 flags ? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Just to irritate Mark ;-) FreeBSD 2.2.6-STABLE #0: Sat Apr 11 18:49:15 GMT 1998 root@hendrix.hallgren.se:/usr/src/sys/compile/HALLGREN CPU: i486 DX2 (486-class CPU) Origin = "GenuineIntel" Id = 0x435 Stepping=5 Features=0x3 real memory = 33554432 (32768K bytes) avail memory = 31223808 (30492K bytes) ... wdc0 at 0x1f0-0x1f7 irq 14 flags 0xff00ff on isa wdc0: unit 0 (wd0): , multi-block-8 wd0: 516MB (1057392 sectors), 1049 cyls, 16 heads, 63 S/T, 512 B/S wdc0: unit 1 (wd1): , multi-block-8 wd1: 1222MB (2503872 sectors), 2484 cyls, 16 heads, 63 S/T, 512 B/S wdc1 at 0x170-0x177 irq 15 flags 0xff00ff on isa wdc1: unit 0 (wd2): , multi-block-16 wd2: 4134MB (8467200 sectors), 8400 cyls, 16 heads, 63 S/T, 512 B/S bash-2.00$ dd if=/dev/rwd1 of=/dev/null bs=1m count=20 20+0 records in 20+0 records out 20971520 bytes transferred in 6.716262 secs (3122499 bytes/sec) bash-2.00$ dd if=/dev/rwd2 of=/dev/null bs=1m count=20 20+0 records in 20+0 records out 20971520 bytes transferred in 6.307651 secs (3324775 bytes/sec) My 486 is running at 66MHz. wdc0 is a dirt cheap ISA multiIO card. wdc1 is a Soundblaster 32(ISA). Of course, the machine freezes while I run these tests. :-/ Tommy Hallgren(md6tommy@mdstud.chalmers.se) - the source of all good beers... Go to http://www.freebsd.org and get real BSD Unix. Today! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 05:34:25 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id FAA15172 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 05:34:25 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from mirage.nlink.com.br (mirage.nlink.com.br [200.238.120.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA15160 for ; Sun, 19 Apr 1998 12:34:20 GMT (envelope-from paulo@nlink.com.br) Received: from localhost (paulo@localhost) by mirage.nlink.com.br (8.8.8/8.8.5) with SMTP id JAA23668 for ; Sun, 19 Apr 1998 09:34:22 -0300 (EST) Date: Sun, 19 Apr 1998 09:34:21 -0300 (EST) From: Paulo Fragoso To: freebsd-hardware@FreeBSD.ORG Subject: Etinc Cards & Riscom Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi All, I'm using a Etinc PCI card with digital link at 64Kbps. Another extrem I'm using a Riscom, that's ok. I've another digital link at 256Kbps with fractional E1 converter (RAD FCD-2L). It work with an IBM 2210 and a Cisco 2501, that's ok. I try to use a Etinc PCI card with a FBSD-2.2.6 and Cisco 2501 in another side. In this configuration I recivie link UP in the FBSD and Line Protocol UP in the Cisco, but don't work!!! In debug mode to Etinc I think that return packets are arriving: ================================================================ Apr 13 12:02:37 tupolev /kernel: PPP/0 IPCP SND CONFIG ACK Apr 13 12:02:37 tupolev /kernel: PPP TX/0 IPCP ff 03 80 21 02 9e 00 0a 03 06 c 0 a8 01 35 Apr 13 12:02:37 tupolev /kernel: PPP RX/0 Prot(8207) 82 07 01 98 00 04 Apr 13 12:02:37 tupolev /kernel: PPP TX/0 LCP ff 03 c0 21 08 1f 00 06 82 07 Apr 13 12:02:37 tupolev /kernel: PPP RX/0 IPCP 80 21 02 00 00 04 Apr 13 12:02:37 tupolev /kernel: PPP/0 Event:IPCP RCV CONFIG ACK New State:CP _OPENED Apr 13 12:02:37 tupolev /kernel: PPP RX/0 IP 00 21 45 00 00 88 96 cb 00 00 3 d 11 dc 07 c8 85 00 24 c8 ee Apr 13 12:02:37 tupolev /kernel: PPP RX/0 IP 00 21 45 00 00 2c 45 4b 40 00 1 b 06 1d 53 c8 f9 f3 40 c8 ee Apr 13 12:02:38 tupolev /kernel: PPP RX/0 IP 00 21 45 00 00 47 70 55 00 00 f 8 11 08 7d c8 dc 40 05 c8 ee Apr 13 12:02:38 tupolev /kernel: PPP RX/0 LCP c0 21 09 01 00 0c 38 be 40 9c 0 0 00 00 00 Apr 13 12:02:38 tupolev /kernel: PPP/0 Event:LCP RCV ECHO New State:CP_OPENE D Apr 13 12:02:38 tupolev /kernel: PPP/0 LCP SND ECHO REPLY seq:1 Apr 13 12:02:38 tupolev /kernel: PPP TX/0 LCP ff 03 c0 21 0a 01 00 0c d4 fc b f ef 00 00 00 00 ================================================================ but 100% of ping packets are lost!!! -Why Etinc card work fine in 64Kbps? -Why Etinc cart don't work in 256Kbps? Is the problem that converter? -Why Cisco and IBM routers work fine in both links (256 and 64)? I try to use in link at 256Kbps my Etinc with in another side a Riscom, it don't work!!! Can anyone help me? Many thanks, Paulo. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 07:10:03 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA24270 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 07:10:03 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from ns.tar.com (ns.tar.com [204.95.187.2]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id OAA24198 for ; Sun, 19 Apr 1998 14:09:57 GMT (envelope-from dick@tar.com) Received: from ppro.tar.com (ppro.tar.com [204.95.187.9]) by ns.tar.com (8.8.8/8.8.7) with SMTP id JAA04135 for ; Sun, 19 Apr 1998 09:09:56 -0500 (CDT) Message-Id: <199804191409.JAA04135@ns.tar.com> From: "Richard Seaman, Jr." To: "hardware@freebsd.org" Date: Sun, 19 Apr 98 09:09:55 -0500 Reply-To: "Richard Seaman, Jr." X-Mailer: PMMail 1.92 For OS/2 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Subject: warning, IDE controller timing not set Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org What's the significance of the warning message: "warning, IDE controller timing not set" Should I be concerned? Thanks for any reply. Sample follows: Apr 18 16:09:54 ns /kernel: Found BIOS32 Service Directory header at 0xf00fb060 Apr 18 16:09:54 ns /kernel: Entry = 0xfb4e0 (0xf00fb4e0) Rev = 0 Len = 1 Apr 18 16:09:54 ns /kernel: PCI BIOS entry at 0xb510 Apr 18 16:09:54 ns /kernel: DMI header at 0xf00f5e30 Apr 18 16:09:54 ns /kernel: Version 2.0 Apr 18 16:09:54 ns /kernel: Table at 0xf0800, 30 entries, 673 bytes Apr 18 16:09:54 ns /kernel: Other BIOS signatures found: Apr 18 16:09:54 ns /kernel: ACPI: 00000000 Apr 18 16:09:54 ns /kernel: $PnP: 000fc080 Apr 18 16:09:54 ns /kernel: pci_open(1): mode 1 addr port (0x0cf8) is 0x80000074 Apr 18 16:09:54 ns /kernel: pci_open(1a): mode1res=0x80000000 (0x80000000) Apr 18 16:09:54 ns /kernel: pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=55911039) Apr 18 16:09:54 ns /kernel: Probing for devices on PCI bus 0: Apr 18 16:09:54 ns /kernel: found-> vendor=0x1039, dev=0x5591, revid=0x02 Apr 18 16:09:54 ns /kernel: class=06-00-00, hdrtype=0x00, mfdev=1 Apr 18 16:09:54 ns /kernel: map[0]: type 1, range 32, base e0000000, size 26 Apr 18 16:09:54 ns /kernel: chip0: rev 0x02 on pci0.0.0 Apr 18 16:09:54 ns /kernel: found-> vendor=0x1039, dev=0x5513, revid=0xd0 Apr 18 16:09:54 ns /kernel: class=01-01-8a, hdrtype=0x00, mfdev=1 Apr 18 16:09:54 ns /kernel: intpin=a, irq=14 Apr 18 16:09:54 ns /kernel: map[0]: type 4, range 32, base 000001f0, size 3 Apr 18 16:09:54 ns /kernel: map[1]: type 4, range 32, base 000003f4, size 2 Apr 18 16:09:54 ns /kernel: map[2]: type 4, range 32, base 00000170, size 3 Apr 18 16:09:54 ns /kernel: map[3]: type 4, range 32, base 00000374, size 2 Apr 18 16:09:54 ns /kernel: map[4]: type 4, range 32, base 00004000, size 4 Apr 18 16:09:54 ns /kernel: ide_pci0: rev 0xd0 int a irq 14 on pci0.0.1 Apr 18 16:09:54 ns /kernel: generic_status: no PCI IDE timing info available Apr 18 16:09:55 ns /kernel: generic_status: no PCI IDE timing info available Apr 18 16:09:55 ns /kernel: ide_pci: busmaster 0 status: 04 from port: 00004002 Apr 18 16:09:55 ns /kernel: generic_status: no PCI IDE timing info available Apr 18 16:09:55 ns /kernel: generic_status: no PCI IDE timing info available Apr 18 16:09:55 ns /kernel: ide_pci: busmaster 1 status: 00 from port: 0000400a Apr 18 16:09:55 ns /kernel: found-> vendor=0x1039, dev=0x0008, revid=0x01 Apr 18 16:09:55 ns /kernel: class=06-01-00, hdrtype=0x00, mfdev=1 Apr 18 16:09:55 ns /kernel: chip1: rev 0x01 on pci0.1.0 Apr 18 16:09:55 ns /kernel: found-> vendor=0x1039, dev=0x0009, revid=0x00 Apr 18 16:09:55 ns /kernel: class=ff-00-00, hdrtype=0x00, mfdev=1 Apr 18 16:09:55 ns /kernel: found-> vendor=0x1039, dev=0x0001, revid=0x00 Apr 18 16:09:55 ns /kernel: class=06-04-00, hdrtype=0x01, mfdev=0 Apr 18 16:09:55 ns /kernel: chip2: rev 0x00 on pci0.2.0 Apr 18 16:09:55 ns /kernel: found-> vendor=0x8086, dev=0x1229, revid=0x04 Apr 18 16:09:55 ns /kernel: class=02-00-00, hdrtype=0x00, mfdev=0 Apr 18 16:09:55 ns /kernel: intpin=a, irq=15 Apr 18 16:09:55 ns /kernel: map[0]: type 3, range 32, base e5101000, size 12 Apr 18 16:09:55 ns /kernel: map[1]: type 4, range 32, base 0000e000, size 5 Apr 18 16:09:55 ns /kernel: map[2]: type 1, range 32, base e5000000, size 20 Apr 18 16:09:56 ns /kernel: wdc0 at 0x1f0-0x1f7 irq 14 flags 0xa0ffa0ff on isa Apr 18 16:09:56 ns /kernel: wdc0: unit 0 (wd0): , 32-bit, multi-block-16 Apr 18 16:09:56 ns /kernel: wd0: 814MB (1667232 sectors), 1654 cyls, 16 heads, 63 S/T, 512 B/S Apr 18 16:09:56 ns /kernel: wd0: ATA INQUIRE valid = 0003, dmamword = 0203, apio = 0001, udma = 0000 Apr 18 16:09:56 ns /kernel: ide_pci: generic_dmainit 01f0:1: warning, IDE controller timing not set Apr 18 16:09:56 ns /kernel: wdc0: unit 1 (wd1): , DMA, 32-bit, multi-block-16 Apr 18 16:09:56 ns /kernel: wd1: 4884MB (10003392 sectors), 9924 cyls, 16 heads, 63 S/T, 512 B/S Apr 18 16:09:56 ns /kernel: wd1: ATA INQUIRE valid = 0003, dmamword = 0407, apio = 0003, udma = 0000 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 11:18:05 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA07760 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 11:18:05 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from mail3.sirius.com (mail3.sirius.com [205.134.253.133]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA07710 for ; Sun, 19 Apr 1998 18:17:55 GMT (envelope-from parag@mail.codegen.com) Received: from [192.168.100.101] (ppp-asok03--134.sirius.net [205.134.244.134]) by mail3.sirius.com (8.8.7/Sirius-8.8.7-97.08.12) with SMTP id LAA22836; Sun, 19 Apr 1998 11:17:48 -0700 (PDT) Message-Id: <199804191817.LAA22836@mail3.sirius.com> Subject: Re: best wdc0 flags ? Date: Sun, 19 Apr 1998 11:19:31 -0700 x-sender: parag@mail.codegen.com x-mailer: Claris Emailer 2.0v3, January 22, 1998 From: Parag Patel To: "Tommy Hallgren" , cc: "Mark Mayo" Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Just for fun, here's what I get on my 486DX-100 (with 32Mb RAM): bash-2.00$ dd if=/dev/rwd0 of=/dev/null bs=1m count=20 20+0 records in 20+0 records out 20971520 bytes transferred in 2.248524 secs (9326794 bytes/sec) bash-2.00$ dd if=/dev/rwd0 of=/dev/null bs=1m count=20 20+0 records in 20+0 records out 20971520 bytes transferred in 2.259819 secs (9280177 bytes/sec) bash-2.00$ dd if=/dev/rwd0 of=/dev/null bs=1m count=20 20+0 records in 20+0 records out 20971520 bytes transferred in 2.248620 secs (9326396 bytes/sec) 'Course, I am cheating. This is a Quantum UDMA Fireball (3Gb) running on a CMD 646U2 PCI-UIDE chip running on 3.0-CURRENT with my driver mods to support the new CMD chip: Clipped from dmesg: [...] ide_pci0: rev 0x05 int a irq 11 on pci0.1 5.0 ide_pci0: adding drives to controller 0: 0 1 2 3 [...] wdc0 at 0x6100-0x6107 flags 0xa0ffa0ff on isa wdc0: unit 0 (wd0): , DMA, 32-bit, multi-block-16 wd0: 3079MB (6306048 sectors), 6256 cyls, 16 heads, 63 S/T, 512 B/S [...] And the rest of the system is most responsive when this test is running. :-) -- Parag Patel To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 20:13:47 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id UAA24467 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 20:13:47 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from quokka.prth.tensor.pgs.com (quokka1.prth.tensor.pgs.com [157.147.224.10]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id DAA24443 for ; Mon, 20 Apr 1998 03:13:30 GMT (envelope-from shocking@ariadne.prth.tensor.pgs.com) Received: from ariadne.tensor.pgs.com (ariadne [157.147.227.36]) by quokka.prth.tensor.pgs.com (8.8.7/8.8.7) with SMTP id LAA11128 for ; Mon, 20 Apr 1998 11:12:43 +0800 (WST) Received: from ariadne by ariadne.tensor.pgs.com (SMI-8.6/SMI-SVR4) id LAA24540; Mon, 20 Apr 1998 11:12:33 +0800 Message-Id: <199804200312.LAA24540@ariadne.tensor.pgs.com> X-Mailer: exmh version 2.0.2 2/24/98 To: hardware@FreeBSD.ORG Subject: Narrow SCSI controllers, and using WIDE drives with them Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 11:12:33 +0800 From: Stephen Hocking-Senior Programmer PGS Tensor Perth Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Is it possible, and if you have one of those 50 pin flat ribbon cables, is it possible to buy adaptors for those D-plug style connectors? Stephen, whose boot barracuda is dying, and is considering IDE Ultra DMA. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 21:00:19 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA06624 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 21:00:19 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from d183-205.uoregon.edu (d183-205.uoregon.edu [128.223.183.205]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id EAA06617 for ; Mon, 20 Apr 1998 04:00:12 GMT (envelope-from gurney_j@efn.org) Received: (from jmg@localhost) by d183-205.uoregon.edu (8.8.7/8.8.7) id UAA20550; Sun, 19 Apr 1998 20:59:57 -0700 (PDT) Message-ID: <19980419205956.41617@hydrogen.nike.efn.org> Date: Sun, 19 Apr 1998 20:59:56 -0700 From: John-Mark Gurney To: Stephen Hocking-Senior Programmer PGS Tensor Perth Cc: hardware@FreeBSD.ORG Subject: Re: Narrow SCSI controllers, and using WIDE drives with them References: <199804200312.LAA24540@ariadne.tensor.pgs.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 In-Reply-To: <199804200312.LAA24540@ariadne.tensor.pgs.com>; from Stephen Hocking-Senior Programmer PGS Tensor Perth on Mon, Apr 20, 1998 at 11:12:33AM +0800 Reply-To: John-Mark Gurney Organization: Cu Networking X-Operating-System: FreeBSD 2.2.1-RELEASE i386 X-PGP-Fingerprint: B7 EC EF F8 AE ED A7 31 96 7A 22 B3 D8 56 36 F4 X-Files: The truth is out there X-URL: http://resnet.uoregon.edu/~gurney_j/ Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Stephen Hocking-Senior Programmer PGS Tensor Perth scribbled this message on Apr 20: > Is it possible, and if you have one of those 50 pin flat ribbon cables, is it > possible to buy adaptors for those D-plug style connectors? yes, this is possible... you can by a ribbon to 50 pin centronix... just drop by your local electronic parts store... they should have 'em... and it works fine... I've done this before I had access to the hd50 to centronix-50 pin cable... -- John-Mark Gurney Modem Rev/FAX: +1 541 346 9237 Cu Networking P.O. Box 5693, 97405 Live in Peace, destroy Micro$oft, support free software, run FreeBSD Don't trust anyone you don't have the source for To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 21:16:13 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA10775 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 21:16:13 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from fly.HiWAAY.net (root@fly.HiWAAY.net [208.147.154.56]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id EAA10739 for ; Mon, 20 Apr 1998 04:16:06 GMT (envelope-from dkelly@nospam.hiwaay.net) Received: from nospam.hiwaay.net (tnt3-163.HiWAAY.net [208.147.146.163]) by fly.HiWAAY.net (8.8.8/8.8.6) with ESMTP id XAA29351; Sun, 19 Apr 1998 23:16:03 -0500 (CDT) Received: from localhost (localhost [127.0.0.1]) by nospam.hiwaay.net (8.8.8/8.8.4) with ESMTP id WAA00526; Sun, 19 Apr 1998 22:55:33 -0500 (CDT) Message-Id: <199804200355.WAA00526@nospam.hiwaay.net> X-Mailer: exmh version 2.0.2 2/24/98 To: Stephen Hocking-Senior Programmer PGS Tensor Perth cc: hardware@FreeBSD.ORG From: David Kelly Subject: Re: Narrow SCSI controllers, and using WIDE drives with them In-reply-to: Message from Stephen Hocking-Senior Programmer PGS Tensor Perth of "Mon, 20 Apr 1998 11:12:33 +0800." <199804200312.LAA24540@ariadne.tensor.pgs.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 19 Apr 1998 22:55:32 -0500 Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Stephen Hocking-Senior Programmer PGS Tensor Perth writes: > > Is it possible, and if you have one of those 50 pin flat ribbon cables, is it > > possible to buy adaptors for those D-plug style connectors? Yes, it is possible to buy adapters. No, it won't work. As a general rule wide HD's will not fall back to a narrow connection. Wide controllers are forced to fall back and handle narrow devices as a special case. The exception (only seen it in documentation, not fact) was a wide 9G IBM drive with SCA connector. There was either a pin on the SCA or an external jumper to force the drive to fall back to narrow SCSI. -- David Kelly N4HHE, dkelly@nospam.hiwaay.net ===================================================================== The human mind ordinarily operates at only ten percent of its capacity -- the rest is overhead for the operating system. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 21:32:15 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA14938 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 21:32:15 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from antipodes.cdrom.com (castles147.castles.com [208.214.165.147]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id EAA14821 for ; Mon, 20 Apr 1998 04:31:59 GMT (envelope-from mike@antipodes.cdrom.com) Received: from antipodes.cdrom.com (localhost [127.0.0.1]) by antipodes.cdrom.com (8.8.8/8.8.5) with ESMTP id VAA03887; Sun, 19 Apr 1998 21:29:11 -0700 (PDT) Message-Id: <199804200429.VAA03887@antipodes.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: Stephen Hocking-Senior Programmer PGS Tensor Perth cc: hardware@FreeBSD.ORG Subject: Re: Narrow SCSI controllers, and using WIDE drives with them In-reply-to: Your message of "Mon, 20 Apr 1998 11:12:33 +0800." <199804200312.LAA24540@ariadne.tensor.pgs.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 19 Apr 1998 21:29:10 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > Is it possible, and if you have one of those 50 pin flat ribbon cables, is it > possible to buy adaptors for those D-plug style connectors? Yes. > Stephen, whose boot barracuda is dying, and is considering IDE Ultra DMA. Even without the "ultra" word, if you only have a single disk, and the machine is a workstation, and you don't plan on swapping heavily, IDE is very effective. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 22:50:59 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA07043 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 22:50:59 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id FAA07014 for ; Mon, 20 Apr 1998 05:50:48 GMT (envelope-from gibbs@narnia.plutotech.com) Received: (from gibbs@localhost) by narnia.plutotech.com (8.8.8/8.7.3) id XAA12945; Sun, 19 Apr 1998 23:46:55 -0600 (MDT) Date: Sun, 19 Apr 1998 23:46:55 -0600 (MDT) From: "Justin T. Gibbs" Message-Id: <199804200546.XAA12945@narnia.plutotech.com> To: David Kelly cc: hardware@FreeBSD.ORG Subject: Re: Narrow SCSI controllers, and using WIDE drives with them Newsgroups: pluto.freebsd.hardware In-Reply-To: <199804200355.WAA00526@nospam.hiwaay.net> User-Agent: tin/pre-1.4-971204 (UNIX) (FreeBSD/3.0-CURRENT (i386)) Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In article <199804200355.WAA00526@nospam.hiwaay.net> you wrote: > Stephen Hocking-Senior Programmer PGS Tensor Perth writes: >> >> Is it possible, and if you have one of those 50 pin flat ribbon cables, is it >> >> possible to buy adaptors for those D-plug style connectors? > > Yes, it is possible to buy adapters. No, it won't work. As a general > rule wide HD's will not fall back to a narrow connection. Wide > controllers are forced to fall back and handle narrow devices as a > special case. You should really read the SCSI spec. All transfers are narrow unless a properly negotiated wide connection is established between target and initiator. On many IBM devices there is a jumper that prevents the device from initiating wide negotiation, but this has no effect on the device's ability to handle 8 bit transfers should a negotiation attempt fail. -- Justin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sun Apr 19 23:11:38 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA10466 for freebsd-hardware-outgoing; Sun, 19 Apr 1998 23:11:38 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id GAA10455 for ; Mon, 20 Apr 1998 06:11:30 GMT (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id BAA00769; Mon, 20 Apr 1998 01:11:02 -0500 (EST) (envelope-from toor) From: "John S. Dyson" Message-Id: <199804200611.BAA00769@dyson.iquest.net> Subject: Re: Narrow SCSI controllers, and using WIDE drives with them In-Reply-To: <199804200429.VAA03887@antipodes.cdrom.com> from Mike Smith at "Apr 19, 98 09:29:10 pm" To: mike@smith.net.au (Mike Smith) Date: Mon, 20 Apr 1998 01:11:01 -0500 (EST) Cc: shocking@prth.pgs.com, hardware@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > > > Is it possible, and if you have one of those 50 pin flat ribbon cables, is it > > possible to buy adaptors for those D-plug style connectors? > > Yes. > > > Stephen, whose boot barracuda is dying, and is considering IDE Ultra DMA. > > Even without the "ultra" word, if you only have a single disk, and the > machine is a workstation, and you don't plan on swapping heavily, IDE > is very effective. > Another caveat (and I am IDE addict, due to cost), is that IDE drives often seem to have infant mortality problems. I like IDE, but have also lost about 1/4 of the drives that I have bought due to early failures. Suggestion: do not trust your drive until you have used it for awhile. John To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 00:07:17 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA20877 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 00:07:17 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from sos.freebsd.dk (sos.freebsd.dk [195.8.129.33]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA20806 for ; Mon, 20 Apr 1998 07:06:48 GMT (envelope-from sos@sos.freebsd.dk) Received: (from sos@localhost) by sos.freebsd.dk (8.8.8/8.8.8) id JAA01103; Mon, 20 Apr 1998 09:05:08 +0200 (MEST) (envelope-from sos) Message-Id: <199804200705.JAA01103@sos.freebsd.dk> Subject: Re: Narrow SCSI controllers, and using WIDE drives with them In-Reply-To: <199804200611.BAA00769@dyson.iquest.net> from "John S. Dyson" at "Apr 20, 98 01:11:01 am" To: toor@dyson.iquest.net (John S. Dyson) Date: Mon, 20 Apr 1998 09:05:08 +0200 (MEST) Cc: mike@smith.net.au, shocking@prth.pgs.com, hardware@FreeBSD.ORG From: Søren Schmidt Reply-to: sos@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL30 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In reply to John S. Dyson who wrote: > > > > Even without the "ultra" word, if you only have a single disk, and the > > machine is a workstation, and you don't plan on swapping heavily, IDE > > is very effective. > > > Another caveat (and I am IDE addict, due to cost), is that IDE drives > often seem to have infant mortality problems. I like IDE, but have also > lost about 1/4 of the drives that I have bought due to early failures. > > Suggestion: do not trust your drive until you have used it for awhile. I can second that one!! Also I have had very bad DOA rates with Quantum disks, but those that work, seems quite reliable... It seems that if the drive survives the first 2-3 weeks of abuse then its virutally impossible to kill :) I have a Maxtor DiamondMax that has been dropped, overheated and what not several times, and it still works like a charm :) -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Søren Schmidt (sos@FreeBSD.org) FreeBSD Core Team Even more code to hack -- will it ever end .. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 00:17:06 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA22464 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 00:17:06 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from math.berkeley.edu (math.Berkeley.EDU [128.32.183.94]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA22455 for ; Mon, 20 Apr 1998 07:17:04 GMT (envelope-from dan@math.berkeley.edu) Received: (from dan@localhost) by math.berkeley.edu (8.8.7/8.8.7) id AAA02451; Mon, 20 Apr 1998 00:17:02 -0700 (PDT) Date: Mon, 20 Apr 1998 00:17:02 -0700 (PDT) From: dan@math.berkeley.edu (Dan Strick) Message-Id: <199804200717.AAA02451@math.berkeley.edu> To: dkelly@hiwaay.net, shocking@prth.pgs.com Subject: Re: Narrow SCSI controllers, and using WIDE drives with them Cc: dan@math.berkeley.edu, hardware@FreeBSD.ORG Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > Is it possible, and if you have one of those 50 pin flat ribbon cables, > > is it possible to buy adaptors for those D-plug style connectors? > > Yes, it is possible to buy adapters. No, it won't work. As a general > rule wide HD's will not fall back to a narrow connection. Wide > controllers are forced to fall back and handle narrow devices as a > special case. I think the intent of the SCSI-2 standard is that wide transfers be attempted only if both SCSI bus nodes and the cable between them are wide and that the required tansfer parameter negotiation should somehow fail otherwise. I don't recall the details. I concede that the SCSI bus manufacturers have in the past demonstrated an unexpected ability for evading obviously desirable features of the SCSI standards when concentrating exclusively on immediate marketing concerns. It ought to work by default. If not, you may be able to tell the host adapter via its BIOS interface to only do narrow transfers with specific SCSI devices. Dan Strick dan@math.berkeley.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 01:01:39 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA00185 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 01:01:39 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dt050n33.san.rr.com (@dt050n33.san.rr.com [204.210.31.51]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id IAA00163; Mon, 20 Apr 1998 08:01:27 GMT (envelope-from Studded@san.rr.com) Received: from san.rr.com (Studded@localhost [127.0.0.1]) by dt050n33.san.rr.com (8.8.8/8.8.8) with ESMTP id BAA18491; Mon, 20 Apr 1998 01:01:21 -0700 (PDT) (envelope-from Studded@san.rr.com) Message-ID: <353B00D0.9A7A03F8@san.rr.com> Date: Mon, 20 Apr 1998 01:01:20 -0700 From: Studded Organization: Triborough Bridge & Tunnel Authority X-Mailer: Mozilla 4.05 [en] (X11; I; FreeBSD 2.2.6-STABLE i386) MIME-Version: 1.0 To: Donald Burr CC: hardware@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: best wdc0 flags ? References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Donald Burr wrote: > The best combination of flags is 0x80ff, this will turn on 32-bit > transfers and will do multi-block up to the maximum that is supported by > the drive. I tried those flags and noticed that it turned on 32-bit transfers for my wd0, resulting in a 20% increase in throughput. Out of curiosity, why aren't these flags included in GENERIC? Doug -- *** Chief Operations Officer, DALnet IRC network *** *** Proud designer and maintainer of the world's largest Internet *** Relay Chat server with 5,328 simultaneous connections. *** Try spider.dal.net on ports 6662-4 (Powered by FreeBSD) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 01:30:38 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA06230 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 01:30:38 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id IAA06171; Mon, 20 Apr 1998 08:30:25 GMT (envelope-from bde@godzilla.zeta.org.au) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.7/8.8.7) id SAA15511; Mon, 20 Apr 1998 18:27:45 +1000 Date: Mon, 20 Apr 1998 18:27:45 +1000 From: Bruce Evans Message-Id: <199804200827.SAA15511@godzilla.zeta.org.au> To: dburr@POBoxes.com, Studded@san.rr.com Subject: Re: best wdc0 flags ? Cc: hardware@FreeBSD.ORG, stable@FreeBSD.ORG Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I tried those flags and noticed that it turned on 32-bit transfers for >my wd0, resulting in a 20% increase in throughput. Out of curiosity, why >aren't these flags included in GENERIC? Becuase they break operation of drives that don't support them. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 04:41:46 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id EAA06291 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 04:41:46 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from gibson.bathl.westnet.net.uk (cyberc.demon.co.uk [193.237.161.19]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA06262 for ; Mon, 20 Apr 1998 11:41:40 GMT (envelope-from graham@cyberc.demon.co.uk) Received: (from graham@localhost) by gibson.bathl.westnet.net.uk (8.8.7/8.6.9) id MAA25693 for hardware@freebsd.org; Mon, 20 Apr 1998 12:40:12 +0100 (BST) From: Graham Breach Message-Id: <199804201140.MAA25693@gibson.bathl.westnet.net.uk> Subject: Smarty and floppy disk driver To: hardware@FreeBSD.ORG Date: Mon, 20 Apr 1998 12:40:12 +0100 (BST) X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi all I have a Fischer Smarty to get working - it's a smart card reader/writer that looks like a floppy disk. The problem I have is that the read and write "sectors" are sectors 100 and 200 on the first track, and I'm not going to get my hands on those using /dev/fd0. I've been looking at the code for the floppy driver and I think my best bet is to add ioctl() for reading and writing to these sectors. Does anyone have any better ideas/suggestions? TIA Graham Breach To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 08:36:10 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id IAA15183 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 08:36:10 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA14855; Mon, 20 Apr 1998 15:34:19 GMT (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id IAA00441; Mon, 20 Apr 1998 08:30:50 -0700 (PDT) Message-Id: <199804201530.IAA00441@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: Bruce Evans cc: dburr@POBoxes.com, Studded@san.rr.com, hardware@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: best wdc0 flags ? In-reply-to: Your message of "Mon, 20 Apr 1998 18:27:45 +1000." <199804200827.SAA15511@godzilla.zeta.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 08:30:49 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > I tried those flags and noticed that it turned on 32-bit transfers for > >my wd0, resulting in a 20% increase in throughput. Out of curiosity, why > >aren't these flags included in GENERIC? > > Becuase they break operation of drives that don't support them. Do we have any examples of controllers that don't? MHO is that at this point we should perhaps consider converting the option to accomodate the majority case, where they do. I've never encountered a controller that didn't work with 32-bit transfers. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 09:11:03 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA24949 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 09:11:03 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA24763; Mon, 20 Apr 1998 16:10:31 GMT (envelope-from bde@godzilla.zeta.org.au) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.7/8.8.7) id CAA04114; Tue, 21 Apr 1998 02:05:03 +1000 Date: Tue, 21 Apr 1998 02:05:03 +1000 From: Bruce Evans Message-Id: <199804201605.CAA04114@godzilla.zeta.org.au> To: bde@zeta.org.au, mike@smith.net.au Subject: Re: best wdc0 flags ? Cc: dburr@POBoxes.com, hardware@FreeBSD.ORG, stable@FreeBSD.ORG, Studded@san.rr.com Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >> > I tried those flags and noticed that it turned on 32-bit transfers for >> >my wd0, resulting in a 20% increase in throughput. Out of curiosity, why >> >aren't these flags included in GENERIC? >> >> Becuase they break operation of drives that don't support them. > >Do we have any examples of controllers that don't? I thought I did, but my oldest accessible drive (all 400MB of it from 4 years ago) supports them. The probe seems to handle any that don't. Setting the multi-block flag is not such a good optimization, since it pessimizes throughput on some drives and it increases interrupt latency. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 09:31:40 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA01430 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 09:31:40 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA01183; Mon, 20 Apr 1998 16:30:32 GMT (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id JAA00670; Mon, 20 Apr 1998 09:26:39 -0700 (PDT) Message-Id: <199804201626.JAA00670@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: Bruce Evans cc: mike@smith.net.au, dburr@POBoxes.com, hardware@FreeBSD.ORG, stable@FreeBSD.ORG, Studded@san.rr.com Subject: Re: best wdc0 flags ? In-reply-to: Your message of "Tue, 21 Apr 1998 02:05:03 +1000." <199804201605.CAA04114@godzilla.zeta.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 09:26:38 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > >> > I tried those flags and noticed that it turned on 32-bit transfers for > >> >my wd0, resulting in a 20% increase in throughput. Out of curiosity, why > >> >aren't these flags included in GENERIC? > >> > >> Becuase they break operation of drives that don't support them. > > > >Do we have any examples of controllers that don't? > > I thought I did, but my oldest accessible drive (all 400MB of it from > 4 years ago) supports them. The probe seems to handle any that don't. OK. Should we make it the default then? > Setting the multi-block flag is not such a good optimization, since it > pessimizes throughput on some drives and it increases interrupt latency. Can you qualify "some drives" again? The overall performance improvement in general use is marked, and it decreases interrupt load in the DMA case. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 09:43:42 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA04639 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 09:43:42 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from caladan.tdx.co.uk (caladan.tdx.co.uk [195.188.177.4]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA04481 for ; Mon, 20 Apr 1998 16:43:12 GMT (envelope-from kpielorz@tdx.co.uk) Received: from tdx.co.uk (lorca-tx.tdx.co.uk [195.188.177.242]) by caladan.tdx.co.uk (8.8.8/8.8.8) with ESMTP id RAA11831; Mon, 20 Apr 1998 17:42:38 +0100 (BST) (envelope-from kpielorz@tdx.co.uk) Message-ID: <353B7AEC.D560BAD7@tdx.co.uk> Date: Mon, 20 Apr 1998 17:42:20 +0100 From: Karl Pielorz Organization: TDX X-Mailer: Mozilla 4.04 [en] (WinNT; I) MIME-Version: 1.0 To: Mike Smith CC: hardware@FreeBSD.ORG Subject: Re: best wdc0 flags ? References: <199804201626.JAA00670@dingo.cdrom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Mike Smith wrote: > > > >> > I tried those flags and noticed that it turned on 32-bit transfers for > > >> >my wd0, resulting in a 20% increase in throughput. Out of curiosity, why > > >> >aren't these flags included in GENERIC? > > >> > > >> Becuase they break operation of drives that don't support them. > > > > > >Do we have any examples of controllers that don't? > > > > I thought I did, but my oldest accessible drive (all 400MB of it from > > 4 years ago) supports them. The probe seems to handle any that don't. > > OK. Should we make it the default then? Not a good idea... 32 bit transfers cause my nice shiny new Fireball 4.3Gb SE's to barf (this is on a 440FX chipset 'embeded' controller). A bit annoying - but probably expected from Quantum... Kp -- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 09:45:34 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA05289 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 09:45:34 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from caladan.tdx.co.uk (caladan.tdx.co.uk [195.188.177.4]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA05211 for ; Mon, 20 Apr 1998 16:45:20 GMT (envelope-from kpielorz@tdx.co.uk) Received: from tdx.co.uk (lorca-tx.tdx.co.uk [195.188.177.242]) by caladan.tdx.co.uk (8.8.8/8.8.8) with ESMTP id RAA11898 for ; Mon, 20 Apr 1998 17:45:16 +0100 (BST) (envelope-from kpielorz@tdx.co.uk) Message-ID: <353B7B8A.65E9B10C@tdx.co.uk> Date: Mon, 20 Apr 1998 17:44:58 +0100 From: Karl Pielorz Organization: TDX X-Mailer: Mozilla 4.04 [en] (WinNT; I) MIME-Version: 1.0 To: freebsd-hardware@FreeBSD.ORG Subject: Recomended ISA SCSI Cards Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Can anyone recomend any good ISA SCSI cards to use with FreeBSD? - Preferably something that'll work with a machine with >16Mb of RAM etc. - I have a 1542 but it's been nothing but trouble... Anything with a 25 way D connector on the back would be great - as it's for an external ZIP drive... I'm not too sure I would want to replace an AHA1542 with an AHA1520/22 though - unless anyone else has any comments? In fact - thinking about it, bootable would also be a big help... (That's probably narrowed it down a bit!) Regards, Karl Pielorz To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 09:56:45 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA08944 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 09:56:45 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA08569; Mon, 20 Apr 1998 16:55:43 GMT (envelope-from bde@godzilla.zeta.org.au) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.7/8.8.7) id CAA05910; Tue, 21 Apr 1998 02:50:39 +1000 Date: Tue, 21 Apr 1998 02:50:39 +1000 From: Bruce Evans Message-Id: <199804201650.CAA05910@godzilla.zeta.org.au> To: bde@zeta.org.au, mike@smith.net.au Subject: Re: best wdc0 flags ? Cc: dburr@POBoxes.com, hardware@FreeBSD.ORG, stable@FreeBSD.ORG, Studded@san.rr.com Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >> I thought I did, but my oldest accessible drive (all 400MB of it from >> 4 years ago) supports them. The probe seems to handle any that don't. > >OK. Should we make it the default then? In -current. >> Setting the multi-block flag is not such a good optimization, since it >> pessimizes throughput on some drives and it increases interrupt latency. > >Can you qualify "some drives" again? The overall performance >improvement in general use is marked, and it decreases interrupt load >in the DMA case. Old drives. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 10:01:40 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA10583 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 10:01:40 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA10436 for ; Mon, 20 Apr 1998 17:01:03 GMT (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id JAA00786; Mon, 20 Apr 1998 09:56:56 -0700 (PDT) Message-Id: <199804201656.JAA00786@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: Karl Pielorz cc: Mike Smith , hardware@FreeBSD.ORG Subject: Re: best wdc0 flags ? In-reply-to: Your message of "Mon, 20 Apr 1998 17:42:20 BST." <353B7AEC.D560BAD7@tdx.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 09:56:56 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Mike Smith wrote: >> >>>>>> I tried those flags and noticed that it turned on 32-bit transfers for >>>>>>my wd0, resulting in a 20% increase in throughput. Out of curiosity, why >>>>>>aren't these flags included in GENERIC? >>>>> >>>>> Becuase they break operation of drives that don't support them. >>>> >>>>Do we have any examples of controllers that don't? >>> >>> I thought I did, but my oldest accessible drive (all 400MB of it from >>> 4 years ago) supports them. The probe seems to handle any that don't. >> >> OK. Should we make it the default then? > > Not a good idea... 32 bit transfers cause my nice shiny new Fireball 4.3Gb > SE's to barf (this is on a 440FX chipset 'embeded' controller). Have you tried enabling DMA? (0xa0ffa0ff) Also check your BIOS configuration to make sure that the disks are set up correctly (PIO mode, etc.) > A bit annoying - but probably expected from Quantum... Not really. They *ought* to work - certainly we're seeing 32-bit transfers working on other SE-series drives just fine. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 10:27:09 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA17420 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 10:27:09 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from ryouko.nas.nasa.gov (ryouko.nas.nasa.gov [129.99.34.103]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA17412 for ; Mon, 20 Apr 1998 17:27:04 GMT (envelope-from greg@ryouko.nas.nasa.gov) Received: from ryouko.nas.nasa.gov (localhost [127.0.0.1]) by ryouko.nas.nasa.gov (8.8.7/NAS8.8.7) with ESMTP id KAA01852; Mon, 20 Apr 1998 10:26:26 -0700 (PDT) Message-Id: <199804201726.KAA01852@ryouko.nas.nasa.gov> To: Karl Pielorz cc: freebsd-hardware@FreeBSD.ORG Subject: Re: Recomended ISA SCSI Cards In-reply-to: Your message of "Mon, 20 Apr 1998 17:44:58 BST." <353B7B8A.65E9B10C@tdx.co.uk> Date: Mon, 20 Apr 1998 10:26:26 -0700 From: "Gregory P. Smith" Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Can anyone recomend any good ISA SCSI cards to use with FreeBSD? - > Preferably something that'll work with a machine with >16Mb of RAM etc. - I > have a 1542 but it's been nothing but trouble... ... > I'm not too sure I would want to replace an AHA1542 with an AHA1520/22 > though - unless anyone else has any comments? "good" and "ISA" should not be used in the same paragraph. Any "good" ISA SCSI card is busmastering. The ISA bus only has 24 address lines, meaning busmastering ISA cards can only access the first 16mb of RAM. Your other option is the port I/O style card such as the AHA1522. These require lots of CPU for the transfer but don't have the 16mb limit. My ISA recommendation is the 1542, followed by a 1522 if you don't like the 16mb thing (really, these are your only sane choices). Most OSes drivers will work around the 16mb limit (w/ a performance hit) and copy data in/out of the low 16mb for busmastering ISA. I don't, however, know the state of FreeBSD's aha1542 driver; they might not have implemented that. Real recommendation: get a PCI motherboard, they're cheap. Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 10:34:53 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA19117 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 10:34:53 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from caladan.tdx.co.uk (caladan.tdx.co.uk [195.188.177.4]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA18987 for ; Mon, 20 Apr 1998 17:34:33 GMT (envelope-from kpielorz@tdx.co.uk) Received: from tdx.co.uk (lorca-tx.tdx.co.uk [195.188.177.242]) by caladan.tdx.co.uk (8.8.8/8.8.8) with ESMTP id SAA12902; Mon, 20 Apr 1998 18:34:20 +0100 (BST) (envelope-from kpielorz@tdx.co.uk) Message-ID: <353B870B.C27F99AD@tdx.co.uk> Date: Mon, 20 Apr 1998 18:34:03 +0100 From: Karl Pielorz Organization: TDX X-Mailer: Mozilla 4.04 [en] (WinNT; I) MIME-Version: 1.0 To: "Gregory P. Smith" CC: freebsd-hardware@FreeBSD.ORG Subject: Re: Recomended ISA SCSI Cards References: <199804201726.KAA01852@ryouko.nas.nasa.gov> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, Gregory P. Smith wrote: > > > Can anyone recomend any good ISA SCSI cards to use with FreeBSD? - > > Preferably something that'll work with a machine with >16Mb of RAM etc. - I > > have a 1542 but it's been nothing but trouble... > ... > > I'm not too sure I would want to replace an AHA1542 with an AHA1520/22 > > though - unless anyone else has any comments? > > "good" and "ISA" should not be used in the same paragraph. I know, I know... Don't remind me! > Any "good" ISA SCSI card is busmastering. The ISA bus only has 24 > address lines, meaning busmastering ISA cards can only access the > first 16mb of RAM. Your other option is the port I/O style card such > as the AHA1522. These require lots of CPU for the transfer but don't > have the 16mb limit. Yep... > My ISA recommendation is the 1542, followed by a 1522 if you don't like > the 16mb thing (really, these are your only sane choices). Most OSes I have a 1542 - but I've seen theres problems with bounce-buffers and stuff on the 1542 - I was looking to moving away from it if I can... > Real recommendation: get a PCI motherboard, they're cheap. The board allready is a PCI motherboard, I've just about got 1 PCI slot free - but I'd rather not waste it on a PCI SCSI card when all it's running is a Zip drive... I think I'll try to find out the current state of the 1542 vs Bounce buffers etc. - if not I'll plumb for the 1522... For once 'speed' is not of the essance... Regards, Karl To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 10:50:02 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA22083 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 10:50:02 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from mrelay.jrc.it (mrelay.jrc.it [139.191.1.65]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA21914; Mon, 20 Apr 1998 17:49:24 GMT (envelope-from nick.hibma@jrc.it) Received: from elect8 (elect8.jrc.it [139.191.71.152]) by mrelay.jrc.it (LMC5688) with SMTP id TAA02205; Mon, 20 Apr 1998 19:49:11 +0200 (MET DST) Date: Mon, 20 Apr 1998 19:46:17 +0200 (MET DST) From: Nick Hibma X-Sender: n_hibma@elect8 Reply-To: Nick Hibma To: freebsd-hackers@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG cc: Mark Willey , Nick Hibma Subject: FreeBSD USB project, help requested Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org The first few steps have been taken in creating a USB device driver stack. First of all, I would like to know if anyone else, beside Mark Willey and me, has also taken on the task of implementing USB support for (Free-)BSD. Second, I would like to hear from people willing to do some testing on the existing software. What has been done: x Implementation of Loadable Kernel Module + 30 lines of PCI driver. x Support for 82371AB chip (also called PIIX4) x UHCI driver (lowest level) x VHUB on top of UHCI, emulates real hub function on host controller x Enumeration of hubs (works great for VHUB :-) x Simple control transfers on the USB bus, based on very simple scheduling, making it possible to do control transfers with real functions. x This resulted in about 5000+ lines of code (documented in C :) What will be done in the near future: x Proper scheduling in UHCI driver with proper retries, interrupt usage through IntOnCompletion, IntOnShortPacket, IntOnError, etc. and implementation of blocking/non-blocking read and write calls rom the USB stack. x Support for 82371Sx chip (also called PIIX/PIIX3) or at least for the version that has USB support. x Support for Isochroneous transfers x Setup of proper interface to applications using USB devices. What we need: x Someone who spends an hour or two getting the snapshot ready for other people to download and try (with a README file telling them what parts of the kernel to patch and how to determine what they have) x People to install the snapshot, run it and send the output of the screen to me. x Programmers interested in taking part in the project by implementing bits and pieces: x Scheduling of UHCI transfers x Implementation of OHCI standard (similar to UHCI) x Support for other UHCI/OHCI chipsets x Support for ISA/EISA/etc. implementations of UHCI x Implementation of device drivers for Interrupt transfer driven devices Everything is rough on the edges and large chunks are missing (like power management support, Suspend/Resume and probably does behave badly on other people's hardware. But without other people's help we won't get there. URL (snapshots): http://www.etla.net/~n_hibma/usb/usb.pl Hope to hear from you. Nick Hibma FreeBSD USB driver development To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 11:12:39 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA27832 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 11:12:39 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from ryouko.nas.nasa.gov (ryouko.nas.nasa.gov [129.99.34.103]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA27554; Mon, 20 Apr 1998 18:11:16 GMT (envelope-from greg@ryouko.nas.nasa.gov) Received: from ryouko.nas.nasa.gov (localhost [127.0.0.1]) by ryouko.nas.nasa.gov (8.8.7/NAS8.8.7) with ESMTP id LAA02256; Mon, 20 Apr 1998 11:10:46 -0700 (PDT) Message-Id: <199804201810.LAA02256@ryouko.nas.nasa.gov> To: Nick Hibma cc: freebsd-hackers@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG, Mark Willey Subject: Re: FreeBSD USB project, help requested In-reply-to: Your message of "Mon, 20 Apr 1998 19:46:17 +0200." Date: Mon, 20 Apr 1998 11:10:46 -0700 From: "Gregory P. Smith" Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > The first few steps have been taken in creating a USB device driver > stack. First of all, I would like to know if anyone else, beside Mark > Willey and me, has also taken on the task of implementing USB support > for (Free-)BSD. Second, I would like to hear from people willing to do > some testing on the existing software. Yes! I have been working on it. I am working on it as a class project (due later this week). I planned on hooking up with your team after I finished what I needed for my project. As it is now, doing as much as I can on my own is a good learning experience for me even if it isn't all used in the end. > x Programmers interested in taking part in the project by implementing > bits and pieces: > x Implementation of OHCI standard (similar to UHCI) > x Support for other UHCI/OHCI chipsets I volunteer for OHCI standard and support for the SiS 5598 OHCI chipset (gee, can you guess what my motherboard has :). > Nick Hibma > FreeBSD USB driver development I look forward to working with you.. Greg Smith To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 11:18:10 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA29475 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 11:18:10 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA29212; Mon, 20 Apr 1998 18:17:11 GMT (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id LAA01068; Mon, 20 Apr 1998 11:13:24 -0700 (PDT) Message-Id: <199804201813.LAA01068@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nick Hibma cc: freebsd-hackers@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG, Mark Willey Subject: Re: FreeBSD USB project, help requested In-reply-to: Your message of "Mon, 20 Apr 1998 19:46:17 +0200." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 11:13:24 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Ah. I wish you'd mentioned you were working on this a while back, or if you did, I wish I'd heard about it! There is a forum for USB developers that's working towards a unified API for USB device drivers, and I think your input would have been greatly appreciated. > The first few steps have been taken in creating a USB device driver > stack. First of all, I would like to know if anyone else, beside Mark > Willey and me, has also taken on the task of implementing USB support > for (Free-)BSD. Second, I would like to hear from people willing to do > some testing on the existing software. I have written some low-level hardware-independent OHCI primitives, enough to probe and initialise the hardware, as well as some basic resource control modules along the lines of the OHCI documents. > What has been done: > > x Implementation of Loadable Kernel Module + 30 lines of PCI driver. > x Support for 82371AB chip (also called PIIX4) Do you only explicitly support this device, or do you support all UHCI implementations? > x UHCI driver (lowest level) > x VHUB on top of UHCI, emulates real hub function on host controller > x Enumeration of hubs (works great for VHUB :-) > x Simple control transfers on the USB bus, based on very > simple scheduling, making it possible to do control transfers with > real functions. > x This resulted in about 5000+ lines of code (documented in C :) This sounds like almost enough for keyboards and mice, and other simple devices. > What will be done in the near future: > > x Proper scheduling in UHCI driver with proper retries, interrupt usage > through IntOnCompletion, IntOnShortPacket, IntOnError, etc. and > implementation of blocking/non-blocking read and write calls rom > the USB stack. > x Support for 82371Sx chip (also called PIIX/PIIX3) or at least for > the version that has USB support. > x Support for Isochroneous transfers > x Setup of proper interface to applications using USB devices. ie. just about everything else. 8) > What we need: > > x Someone who spends an hour or two getting the snapshot ready for other > people to download and try (with a README file telling them what parts > of the kernel to patch and how to determine what they have) You really need to do that. 8) > x People to install the snapshot, run it and send the output of the > screen to me. Gladly, just as soon as I get my hands on some USB peripherals! > x Programmers interested in taking part in the project by implementing > bits and pieces: > x Scheduling of UHCI transfers > x Implementation of OHCI standard (similar to UHCI) > x Support for other UHCI/OHCI chipsets You should stick to the register interfaces as documented in the [OU]HCI standards - this will let you work on any compliant chipset, modulo quirks. > x Support for ISA/EISA/etc. implementations of UHCI Do any of these exist in a documented form? > Hope to hear from you. Oh, you will! -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 11:21:47 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA00758 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 11:21:47 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA00536; Mon, 20 Apr 1998 18:20:46 GMT (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id LAA01099; Mon, 20 Apr 1998 11:17:07 -0700 (PDT) Message-Id: <199804201817.LAA01099@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: "Gregory P. Smith" cc: Nick Hibma , freebsd-hackers@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG, Mark Willey Subject: Re: FreeBSD USB project, help requested In-reply-to: Your message of "Mon, 20 Apr 1998 11:10:46 PDT." <199804201810.LAA02256@ryouko.nas.nasa.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 11:17:07 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > x Programmers interested in taking part in the project by implementing > > bits and pieces: > > x Implementation of OHCI standard (similar to UHCI) > > x Support for other UHCI/OHCI chipsets > > I volunteer for OHCI standard and support for the SiS 5598 OHCI chipset > (gee, can you guess what my motherboard has :). Get the OHCI standard and work from that. (Which is what I have done for my unknown Toshiba chipset.) The whole idea behind OHCI is that you don't have to know what sort of chipset you have, and we shouldn't care at all. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 11:22:27 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA01020 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 11:22:27 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from george.arc.nasa.gov (george.arc.nasa.gov [128.102.194.142]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA00802 for ; Mon, 20 Apr 1998 18:21:43 GMT (envelope-from lamaster@george.arc.nasa.gov) Received: (from lamaster@localhost) by george.arc.nasa.gov (8.8.7/8.8.7) id LAA02800; Mon, 20 Apr 1998 11:21:28 -0700 (PDT) Date: Mon, 20 Apr 1998 11:21:28 -0700 (PDT) From: Hugh LaMaster To: freebsd-hardware@FreeBSD.ORG Subject: BX-based boards [was Re: Dual PP MB of choice?] In-Reply-To: <2.2.32.19980409161524.00cee1ec@gw1> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Anybody tried the new BX-based boards? I see both Intel and Supermicro boards being offered for sale, although I don't know how long you would have to wait in line to actually take delivery. I'm curious because I have an application which appreciates higher memory bandwidth, and the BX boards may (or may not - I haven't seen any test results) have greater bandwidth, especially if you run the memory bus at 100 MHz. [8ns "PC100" SDRAM required for this]. Anybody running FreeBSD on such a board yet? Any gotchas? -- Hugh LaMaster, M/S 233-21, ASCII Email: hlamaster@mail.arc.nasa.gov NASA Ames Research Center Or: lamaster@george.arc.nasa.gov Moffett Field, CA 94035-1000 No Junkmail: USC 18 section 2701 Phone: 650/604-1056 Disclaimer: Unofficial, personal *opinion*. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 11:25:39 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA02482 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 11:25:39 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from mrelay.jrc.it (mrelay.jrc.it [139.191.1.65]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA02269; Mon, 20 Apr 1998 18:25:06 GMT (envelope-from nick.hibma@jrc.it) Received: from elect8 (elect8.jrc.it [139.191.71.152]) by mrelay.jrc.it (LMC5688) with SMTP id UAA02679; Mon, 20 Apr 1998 20:24:48 +0200 (MET DST) Date: Mon, 20 Apr 1998 20:21:57 +0200 (MET DST) From: Nick Hibma X-Sender: n_hibma@elect8 Reply-To: Nick Hibma To: Mike Smith cc: "Gregory P. Smith" , freebsd-hackers@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG, Mark Willey Subject: Re: FreeBSD USB project, help requested In-Reply-To: <199804201817.LAA01099@dingo.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > > x Programmers interested in taking part in the project by implementing > > > bits and pieces: > > > x Implementation of OHCI standard (similar to UHCI) > > > x Support for other UHCI/OHCI chipsets > > > > I volunteer for OHCI standard and support for the SiS 5598 OHCI chipset > > (gee, can you guess what my motherboard has :). > > Get the OHCI standard and work from that. (Which is what I have done > for my unknown Toshiba chipset.) The whole idea behind OHCI is that > you don't have to know what sort of chipset you have, and we shouldn't > care at all. Same for UHCI spec. Be aware of the fact though that without support for separate chipsets you will not be able to figure out the way to do it according to the specs. The final result should be according to [OU]HCI spec and chipset independent, I agree, but for the moment we do not have the knowledge nor the base to claim that it is chipset independent. Nick STA-ISIS, Joint Research Centre, Italy building: 27A tel.: +39 332 78 9549 fax.: +39 332 78 9185 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 11:44:07 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA03917 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 11:29:35 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from sos.freebsd.dk (sos.freebsd.dk [195.8.129.33]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA03646 for ; Mon, 20 Apr 1998 18:28:45 GMT (envelope-from sos@sos.freebsd.dk) Received: (from sos@localhost) by sos.freebsd.dk (8.8.8/8.8.8) id UAA01331; Mon, 20 Apr 1998 20:27:45 +0200 (MEST) (envelope-from sos) Message-Id: <199804201827.UAA01331@sos.freebsd.dk> Subject: Re: best wdc0 flags ? In-Reply-To: <199804201656.JAA00786@dingo.cdrom.com> from Mike Smith at "Apr 20, 98 09:56:56 am" To: mike@smith.net.au (Mike Smith) Date: Mon, 20 Apr 1998 20:27:45 +0200 (MEST) Cc: kpielorz@tdx.co.uk, mike@smith.net.au, hardware@FreeBSD.ORG From: Søren Schmidt Reply-to: sos@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL30 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In reply to Mike Smith who wrote: > > Not a good idea... 32 bit transfers cause my nice shiny new Fireball 4.3Gb > > SE's to barf (this is on a 440FX chipset 'embeded' controller). > > Have you tried enabling DMA? (0xa0ffa0ff) Also check your BIOS > configuration to make sure that the disks are set up correctly (PIO > mode, etc.) Ahem, I've seen this on ALL quantum drives and the natoma chipset.... The natoma apparently pushes the timing to the limit, and crappy drives can't keep up (btw all my Maxtors works just fine)... I bet that you use autoconfig for the drives and that it picks pio4 mode for the Quantum. Then go into the BIOS setup and set it to use pio3 mode hard. And voila it works again.... > > A bit annoying - but probably expected from Quantum... > > Not really. They *ought* to work - certainly we're seeing 32-bit > transfers working on other SE-series drives just fine. No this is expected from Quantum, they have earned plenty of minusses in my book lately (DOA's, sub spec ratings, fastageing etc etc)... -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Søren Schmidt (sos@FreeBSD.org) FreeBSD Core Team Even more code to hack -- will it ever end .. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 11:54:02 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA11556 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 11:54:02 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA11455; Mon, 20 Apr 1998 18:53:54 GMT (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id LAA01211; Mon, 20 Apr 1998 11:50:57 -0700 (PDT) Message-Id: <199804201850.LAA01211@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: sos@FreeBSD.ORG cc: mike@smith.net.au (Mike Smith), kpielorz@tdx.co.uk, hardware@FreeBSD.ORG Subject: Re: best wdc0 flags ? In-reply-to: Your message of "Mon, 20 Apr 1998 20:27:45 +0200." <199804201827.UAA01331@sos.freebsd.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Date: Mon, 20 Apr 1998 11:50:56 -0700 From: Mike Smith Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by hub.freebsd.org id SAA11458 Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > In reply to Mike Smith who wrote: > > > Not a good idea... 32 bit transfers cause my nice shiny new Fireball 4.3Gb > > > SE's to barf (this is on a 440FX chipset 'embeded' controller). > > > > Have you tried enabling DMA? (0xa0ffa0ff) Also check your BIOS > > configuration to make sure that the disks are set up correctly (PIO > > mode, etc.) > > Ahem, I've seen this on ALL quantum drives and the natoma chipset.... > The natoma apparently pushes the timing to the limit, and crappy > drives can't keep up (btw all my Maxtors works just fine)... > > I bet that you use autoconfig for the drives and that it picks pio4 > mode for the Quantum. Then go into the BIOS setup and set it to > use pio3 mode hard. And voila it works again.... If we can confirm this, then it would be wise to add a quirk entry for the Quantums, and go forward with the 32-bit transfer as default. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 12:13:03 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA17272 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 12:13:03 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA17194; Mon, 20 Apr 1998 19:12:43 GMT (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id MAA01295; Mon, 20 Apr 1998 12:07:47 -0700 (PDT) Message-Id: <199804201907.MAA01295@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nick Hibma cc: Mike Smith , "Gregory P. Smith" , freebsd-hackers@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG, Mark Willey Subject: Re: FreeBSD USB project, help requested In-reply-to: Your message of "Mon, 20 Apr 1998 20:21:57 +0200." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 12:07:47 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > > > x Programmers interested in taking part in the project by implementing > > > > bits and pieces: > > > > x Implementation of OHCI standard (similar to UHCI) > > > > x Support for other UHCI/OHCI chipsets > > > > > > I volunteer for OHCI standard and support for the SiS 5598 OHCI chipset > > > (gee, can you guess what my motherboard has :). > > > > Get the OHCI standard and work from that. (Which is what I have done > > for my unknown Toshiba chipset.) The whole idea behind OHCI is that > > you don't have to know what sort of chipset you have, and we shouldn't > > care at all. > > Same for UHCI spec. Be aware of the fact though that without support for > separate chipsets you will not be able to figure out the way to do it > according to the specs. I'm not sure I follow you here. You can locate [OU]HCI devices by their PCI class/subclass codes without needing to resort to vendor/product IDs. > The final result should be according to [OU]HCI spec and chipset > independent, I agree, but for the moment we do not have the knowledge > nor the base to claim that it is chipset independent. If we cleave to the standard to start with, we'll be better off than trying to separate the standard-compliant parts from the chipset-specific parts later on... -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 12:33:34 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA22553 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 12:33:34 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from smtp3.xs4all.nl (smtp3.xs4all.nl [194.109.6.53]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA22397 for ; Mon, 20 Apr 1998 19:32:05 GMT (envelope-from schofiel@xs4all.nl) Received: from xs2.xs4all.nl (root@xs2.xs4all.nl [194.109.6.43]) by smtp3.xs4all.nl (8.8.8/8.8.8) with ESMTP id VAA22428 for ; Mon, 20 Apr 1998 21:31:45 +0200 (CEST) Received: from excelsior (enterprise.xs4all.nl [194.109.14.215]) by xs2.xs4all.nl (8.8.8/8.8.6) with SMTP id VAA29525 for ; Mon, 20 Apr 1998 21:28:59 +0200 (MET DST) Message-ID: <353BA14D.1F49@xs4all.nl> Date: Mon, 20 Apr 1998 21:26:05 +0200 From: Rob Schofield Reply-To: schofiel@xs4all.nl Organization: Knights of the Round Table, Ltd. X-Mailer: Mozilla 3.04Gold (WinNT; I) MIME-Version: 1.0 To: freebsd-hardware@FreeBSD.ORG Subject: Re: Recomended ISA SCSI Cards References: <199804201726.KAA01852@ryouko.nas.nasa.gov> <353B870B.C27F99AD@tdx.co.uk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Karl Pielorz wrote: > I have a 1542 - but I've seen theres problems with bounce-buffers and stuff > on the 1542 - I was looking to moving away from it if I can... This is not applicable if you are talking about a motherboard with a shared PCI/ISA bus bridge setup, and a decent bios. What version of the 1542 are you talking here, A, B or C? Or OEM flavours? > > > Real recommendation: get a PCI motherboard, they're cheap. > > The board allready is a PCI motherboard, I've just about got 1 PCI slot free > - but I'd rather not waste it on a PCI SCSI card when all it's running is a > Zip drive... Err... to be honest, you're probably gonna spend more money on buying (and time setting up) a SCSI card, plus the in-line terminator you'll need, and the type A SCSI 1 -> Apple DB25 connector adaptor. Surely you are better off selling on the ZIP and buying either the parallel port version (setup in seconds, uses a printer cable) or going for the ZIP Plus+ (Uuurgghh who thought that one up) drive, which gives you best of both worlds. And yes, since you ask, I have just done *EXACTLY* that - spot the idiot speaking from experience... Rob Schofield To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 12:46:44 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA25978 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 12:46:44 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA25907; Mon, 20 Apr 1998 19:46:13 GMT (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id OAA02812; Mon, 20 Apr 1998 14:44:05 -0500 (EST) (envelope-from toor) From: "John S. Dyson" Message-Id: <199804201944.OAA02812@dyson.iquest.net> Subject: Re: best wdc0 flags ? In-Reply-To: <199804201605.CAA04114@godzilla.zeta.org.au> from Bruce Evans at "Apr 21, 98 02:05:03 am" To: bde@zeta.org.au (Bruce Evans) Date: Mon, 20 Apr 1998 14:44:05 -0500 (EST) Cc: bde@zeta.org.au, mike@smith.net.au, dburr@POBoxes.com, hardware@FreeBSD.ORG, stable@FreeBSD.ORG, Studded@san.rr.com X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > >> > I tried those flags and noticed that it turned on 32-bit transfers for > >> >my wd0, resulting in a 20% increase in throughput. Out of curiosity, why > >> >aren't these flags included in GENERIC? > >> > >> Becuase they break operation of drives that don't support them. > > > >Do we have any examples of controllers that don't? > > I thought I did, but my oldest accessible drive (all 400MB of it from > 4 years ago) supports them. The probe seems to handle any that don't. > > Setting the multi-block flag is not such a good optimization, since it > pessimizes throughput on some drives and it increases interrupt latency. > What about defaulting to multi-block 4 instead of maximum. This would help the latency issue (re: 16), and mitigate alot of the CPU performance issues (re: 1)? John To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 13:05:01 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA01683 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 13:05:01 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from caladan.tdx.co.uk (caladan.tdx.co.uk [195.188.177.4]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA01545 for ; Mon, 20 Apr 1998 20:04:31 GMT (envelope-from kpielorz@tdx.co.uk) Received: from tdx.co.uk (lorca-tx.tdx.co.uk [195.188.177.242]) by caladan.tdx.co.uk (8.8.8/8.8.8) with ESMTP id VAA15695; Mon, 20 Apr 1998 21:04:05 +0100 (BST) (envelope-from kpielorz@tdx.co.uk) Message-ID: <353BAA24.61AF16EC@tdx.co.uk> Date: Mon, 20 Apr 1998 21:03:48 +0100 From: Karl Pielorz Organization: TDX X-Mailer: Mozilla 4.04 [en] (WinNT; I) MIME-Version: 1.0 To: Mike Smith CC: hardware@FreeBSD.ORG Subject: Re: best wdc0 flags ? References: <199804201850.LAA01211@dingo.cdrom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Mike Smith wrote: > > The natoma apparently pushes the timing to the limit, and crappy > > drives can't keep up (btw all my Maxtors works just fine)... > > If we can confirm this, then it would be wise to add a quirk entry for > the Quantums, and go forward with the 32-bit transfer as default. Thus far they are working OK under 3.0-CURRENT (as of 15th(ish) of April) - and I'm using "flags 0xa0ffa0ff", which I beleive is bus-mastering 32 bit "max. multisector" the drive will negotiate for... They survived a 5gb 'restore' from tape, and a 3gig dump to another SCSI drive... If I put them back in my 2.2.6 system and go for 32 bit / multi-sector - they barf... If I run them on 2.2.6 with just multi-sector - there happy... I guess this doesn't really help much :-) Both my 2.2.6 and 3.0 systems are running 440FX chipsets... I guess if were talking defaults for 3.0 I'd say 32 bit is probably OK... Kp -- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 13:14:51 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA04984 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 13:14:51 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA04937 for ; Mon, 20 Apr 1998 20:14:33 GMT (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id NAA01631; Mon, 20 Apr 1998 13:10:24 -0700 (PDT) Message-Id: <199804202010.NAA01631@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: Karl Pielorz cc: Mike Smith , hardware@FreeBSD.ORG Subject: Re: best wdc0 flags ? In-reply-to: Your message of "Mon, 20 Apr 1998 21:03:48 BST." <353BAA24.61AF16EC@tdx.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 13:10:23 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Mike Smith wrote: > > > > The natoma apparently pushes the timing to the limit, and crappy > > > drives can't keep up (btw all my Maxtors works just fine)... > > > > If we can confirm this, then it would be wise to add a quirk entry for > > the Quantums, and go forward with the 32-bit transfer as default. > > Thus far they are working OK under 3.0-CURRENT (as of 15th(ish) of April) - > and I'm using "flags 0xa0ffa0ff", which I beleive is bus-mastering 32 bit > "max. multisector" the drive will negotiate for... Using DMA avoids the whole 32-bit-transfer issue. (No programmed I/O) > If I put them back in my 2.2.6 system and go for 32 bit / multi-sector - > they barf... Try taking Soren's advice and cutting your BIOS back to mode3 PIO. He has actual experience with these drives, so I'd be inclined to take his word for it. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 13:22:20 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA06988 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 13:22:20 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA06767 for ; Mon, 20 Apr 1998 20:21:21 GMT (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id PAA03068; Mon, 20 Apr 1998 15:20:32 -0500 (EST) (envelope-from toor) From: "John S. Dyson" Message-Id: <199804202020.PAA03068@dyson.iquest.net> Subject: Re: best wdc0 flags ? In-Reply-To: <199804202010.NAA01631@dingo.cdrom.com> from Mike Smith at "Apr 20, 98 01:10:23 pm" To: mike@smith.net.au (Mike Smith) Date: Mon, 20 Apr 1998 15:20:32 -0500 (EST) Cc: kpielorz@tdx.co.uk, mike@smith.net.au, hardware@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > Mike Smith wrote: > > > > > > The natoma apparently pushes the timing to the limit, and crappy > > > > drives can't keep up (btw all my Maxtors works just fine)... > > > > > > If we can confirm this, then it would be wise to add a quirk entry for > > > the Quantums, and go forward with the 32-bit transfer as default. > > > > Thus far they are working OK under 3.0-CURRENT (as of 15th(ish) of April) - > > and I'm using "flags 0xa0ffa0ff", which I beleive is bus-mastering 32 bit > > "max. multisector" the drive will negotiate for... > > Using DMA avoids the whole 32-bit-transfer issue. (No programmed I/O) > Oh, that's right. The multi-block mode ends up being superfluous also. John To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 13:54:41 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA14943 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 13:54:41 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from kaori.communique.net (kaori.communique.net [204.27.67.55]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA14878 for ; Mon, 20 Apr 1998 20:54:29 GMT (envelope-from rzig@verio.net) Received: by kaori.communique.net with Internet Mail Service (5.0.1458.49) id ; Mon, 20 Apr 1998 15:53:26 -0500 Message-ID: From: Raul Zighelboim To: "'hardware@freebsd.org'" Subject: which multiport serial controler ? Date: Mon, 20 Apr 1998 15:53:24 -0500 X-Priority: 3 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.0.1458.49) Content-Type: text/plain Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello there; I am shopping for a multiport serial controller (at least 8 ports) to set up a 'console server', ie, to connect the serial console of a serie of servers. This system will run FreeBSd, so ... which of the supported controller is best suted for the task ? Thanks ================================================== Raul Zighelboim rzig@verio.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 14:22:04 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA23390 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 14:22:04 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from pegasus.com (pegasus.com [206.127.225.31]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id VAA23155 for ; Mon, 20 Apr 1998 21:20:50 GMT (envelope-from richard@pegasus.com) Received: by pegasus.com (8.6.8/PEGASUS-2.2) id LAA25595; Mon, 20 Apr 1998 11:00:29 -1000 Date: Mon, 20 Apr 1998 11:00:29 -1000 From: richard@pegasus.com (Richard Foulk) Message-Id: <199804202100.LAA25595@pegasus.com> X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: hardware@FreeBSD.ORG Subject: PAO w/SSFDC pccard? Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Aloha, I asked recently about connecting a PC-Card interface to an SSFDC (SmartMedia eeprom card, for Olympus digital camera) and someone said check out PAO ... Yikes! Okay, there's no entry in pccard.conf.sample that exactly matches my card (several are close). But even if there was I don't understand what the FAQ is talking about (in the Flash ATA card section) when it says: (this example assumes the Flash ATA disk controller was attached as "wdc1", and the Flash ATA disk is attached as "wd1", as defined in the PAO config file and pccard.conf.sample) How is all this attaching done? With `pccardc enabler' perhaps? How? Before I jump in and start reverse-engineering the code, perhaps someone can point me at some documentation I missed or provide the necessary magic. I've compiled all the junk and patched and rebuilt the kernel which now reports at boot: crd: ctlr(0) Cirrus Logic PD672X (5 mem & 2 I/O windows) pcic: controller I/O address 0x3e0 pcic: controller irq 5 (shared) pcic: controller I/O address 0x3e0 This is on a desktop machine with a pccard adaptor. And a few peeks with pccardc shows: ms# pccardc dumpcis 2 slots found ms# pccardc rdmap Mem 0: flags 0x000 host 0x0 card 0000 size 0 bytes Mem 1: flags 0x000 host 0x0 card 0000 size 0 bytes Mem 2: flags 0x000 host 0x0 card 0000 size 0 bytes Mem 3: flags 0x000 host 0x0 card 0000 size 0 bytes Mem 4: flags 0x000 host 0x0 card 0000 size 0 bytes I/O 0: flags 0x000 port 0x 0 size 0 bytes I/O 1: flags 0x000 port 0x 0 size 0 bytes Mem 0: flags 0x000 host 0x0 card 0000 size 0 bytes Mem 1: flags 0x000 host 0x0 card 0000 size 0 bytes Mem 2: flags 0x000 host 0x0 card 0000 size 0 bytes Mem 3: flags 0x000 host 0x0 card 0000 size 0 bytes Mem 4: flags 0x000 host 0x0 card 0000 size 0 bytes I/O 0: flags 0x000 port 0x 0 size 0 bytes I/O 1: flags 0x000 port 0x 0 size 0 bytes ms# pccardc rdreg Registers for slot 0 00: 82 73 10 00 00 5f 00 00 00 00 00 00 00 00 00 00 10: 00 00 00 00 00 00 11 00 00 00 00 00 00 00 00 f8 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 81 00 30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Registers for slot 1 00: 82 73 10 00 00 5f 00 00 00 00 00 00 00 00 00 00 10: 00 00 00 00 00 00 11 00 00 00 00 00 00 00 00 f8 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 81 00 30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ms# Thanks Richard To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 14:36:58 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA27569 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 14:36:58 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from pegasus.com (pegasus.com [206.127.225.31]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id VAA27239 for ; Mon, 20 Apr 1998 21:35:46 GMT (envelope-from richard@pegasus.com) Received: by pegasus.com (8.6.8/PEGASUS-2.2) id LAA25781; Mon, 20 Apr 1998 11:35:18 -1000 Date: Mon, 20 Apr 1998 11:35:18 -1000 From: richard@pegasus.com (Richard Foulk) Message-Id: <199804202135.LAA25781@pegasus.com> X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: hardware@FreeBSD.ORG Subject: JVC XR-W2010 CD-ROM burner? Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Has anyone used a JVC XR-W2010 CD-ROM burner with FreeBSD? Successfully? Richard To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 14:56:08 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA02821 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 14:56:08 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from caladan.tdx.co.uk (caladan.tdx.co.uk [195.188.177.4]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA02741 for ; Mon, 20 Apr 1998 21:55:59 GMT (envelope-from kpielorz@tdx.co.uk) Received: from tdx.co.uk (lorca-tx.tdx.co.uk [195.188.177.242]) by caladan.tdx.co.uk (8.8.8/8.8.8) with ESMTP id WAA17770; Mon, 20 Apr 1998 22:55:46 +0100 (BST) (envelope-from kpielorz@tdx.co.uk) Message-ID: <353BC450.D30240AB@tdx.co.uk> Date: Mon, 20 Apr 1998 22:55:28 +0100 From: Karl Pielorz Organization: TDX X-Mailer: Mozilla 4.04 [en] (WinNT; I) MIME-Version: 1.0 To: Mike Smith CC: hardware@FreeBSD.ORG Subject: Re: best wdc0 flags ? References: <199804202010.NAA01631@dingo.cdrom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Mike Smith wrote: > Try taking Soren's advice and cutting your BIOS back to mode3 PIO. He > has actual experience with these drives, so I'd be inclined to take his > word for it. I thought the BIOS config was overwritten by the wd device as it came up? - i.e. setup every time? - In fact I thought a lot of the BIOS stuff was just ignored anyway - and rediscovered at probe time or overwritten by the Kernel (e.g. floppy drive type etc.) Of course I do have one complicating factor... None of the drives appear in the BIOS - if I set them in the BIOS to anything else other than 'Not installed' the system won't boot off my SCSI drive (sd0) - the IDE's were an afterthought (and the BIOS doesn't have a 'Boot from SCSI' option :-( I could put the loader etc. on one of the IDE's, but I'd rather not at this point... If DMA is avoiding all the problems I'm probably going to leave it be... I have another machine to setup soon - pure IDE - and suprise, suprise - it's also getting a 4.3Gb SE - I'll try out the different BIOS options on that... (It's also 440FX)... Regards, Karl To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 14:59:09 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA03783 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 14:59:09 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from gatekeeper.alcatel.com.au (gatekeeper.alcatel.com.au [203.17.66.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA03623 for ; Mon, 20 Apr 1998 21:58:53 GMT (envelope-from Peter.Jeremy@alcatel.com.au) Received: from mfg1.cim.alcatel.com.au ([139.188.23.1]) by gatekeeper.alcatel.com.au (PMDF V5.1-7 #U2695) with ESMTP id <01IW4GWYWJVK000A41@gatekeeper.alcatel.com.au> for freebsd-hardware@freebsd.org; Tue, 21 Apr 1998 07:58:20 +1000 Received: from cbd.alcatel.com.au by cim.alcatel.com.au (PMDF V5.1-10 #U2695) with ESMTP id <01IW4GTP3VKWDDYTQW@cim.alcatel.com.au> for freebsd-hardware@freebsd.org; Tue, 21 Apr 1998 07:57:59 +1000 Received: from gsms01.alcatel.com.au by cbd.alcatel.com.au (PMDF V5.1-7 #U2695) with ESMTP id <01IW4GHZ2K1CAZTR97@cbd.alcatel.com.au> for freebsd-hardware@freebsd.org; Tue, 21 Apr 1998 07:46:15 +1100 Received: (from jeremyp@localhost) by gsms01.alcatel.com.au (8.8.8/8.7.3) id HAA29589 for freebsd-hardware@freebsd.org; Tue, 21 Apr 1998 07:46:13 +1000 (EST) Date: Tue, 21 Apr 1998 07:46:13 +1000 (EST) From: Peter Jeremy Subject: Compaq Deskpro 6000 SCSI controller To: freebsd-hardware@FreeBSD.ORG Message-id: <199804202146.HAA29589@gsms01.alcatel.com.au> Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org For reasons beyond my control, I may be forced to get a Deskpro 6000. Does anyone know if the built-in SCSI controller is supported by FreeBSD? I know the integrated NIC isn't supported. Are there any other gotcha's that I should be aware of? Peter -- Peter Jeremy (VK2PJ) peter.jeremy@alcatel.com.au Alcatel Australia Limited 41 Mandible St Phone: +61 2 9690 5019 ALEXANDRIA NSW 2015 Fax: +61 2 9690 5247 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 16:37:59 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA26235 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 16:37:59 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from fly.HiWAAY.net (fly.HiWAAY.net [208.147.154.56]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA26056 for ; Mon, 20 Apr 1998 23:37:00 GMT (envelope-from dkelly@nospam.hiwaay.net) Received: from nospam.hiwaay.net (tnt1-36.HiWAAY.net [208.147.147.36]) by fly.HiWAAY.net (8.8.8/8.8.6) with ESMTP id SAA32307; Mon, 20 Apr 1998 18:36:15 -0500 (CDT) Received: from localhost (localhost [127.0.0.1]) by nospam.hiwaay.net (8.8.8/8.8.4) with ESMTP id SAA04376; Mon, 20 Apr 1998 18:36:13 -0500 (CDT) Message-Id: <199804202336.SAA04376@nospam.hiwaay.net> X-Mailer: exmh version 2.0.2 2/24/98 To: "Justin T. Gibbs" Cc: hardware@FreeBSD.ORG From: David Kelly Subject: Re: Narrow SCSI controllers, and using WIDE drives with them In-reply-to: Message from "Justin T. Gibbs" of "Sun, 19 Apr 1998 23:46:55 MDT." <199804200546.XAA12945@narnia.plutotech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 18:36:13 -0500 Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org "Justin T. Gibbs" writes: > > > > Yes, it is possible to buy adapters. No, it won't work. As a general > > rule wide HD's will not fall back to a narrow connection. Wide > > controllers are forced to fall back and handle narrow devices as a > > special case. > > You should really read the SCSI spec. All transfers are narrow > unless a properly negotiated wide connection is established between > target and initiator. On many IBM devices there is a jumper that > prevents the device from initiating wide negotiation, but this has > no effect on the device's ability to handle 8 bit transfers should > a negotiation attempt fail. I shouldn't have unconditionally stated a wide HD will not work on a narrow SCSI controller. I should have listed the exact model wide Seagate 2G drive (it was a Sun OEM, I could probably find the drive if it matters) that would not work on my 2.2.5 system with Adaptec 2940. Actually, it never got as far as BootEasy. The drive was probed and ID'd but negotiations failed there and the system hung. -- David Kelly N4HHE, dkelly@nospam.hiwaay.net ===================================================================== The human mind ordinarily operates at only ten percent of its capacity -- the rest is overhead for the operating system. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 16:40:37 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA26948 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 16:40:37 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from fly.HiWAAY.net (root@fly.HiWAAY.net [208.147.154.56]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA26755 for ; Mon, 20 Apr 1998 23:39:37 GMT (envelope-from dkelly@nospam.hiwaay.net) Received: from nospam.hiwaay.net (tnt1-36.HiWAAY.net [208.147.147.36]) by fly.HiWAAY.net (8.8.8/8.8.6) with ESMTP id SAA12436; Mon, 20 Apr 1998 18:39:33 -0500 (CDT) Received: from localhost (localhost [127.0.0.1]) by nospam.hiwaay.net (8.8.8/8.8.4) with ESMTP id SAA04391; Mon, 20 Apr 1998 18:39:31 -0500 (CDT) Message-Id: <199804202339.SAA04391@nospam.hiwaay.net> X-Mailer: exmh version 2.0.2 2/24/98 To: dan@math.berkeley.edu (Dan Strick) cc: dkelly@hiwaay.net, shocking@prth.pgs.com, hardware@FreeBSD.ORG From: David Kelly Subject: Re: Narrow SCSI controllers, and using WIDE drives with them In-reply-to: Message from dan@math.berkeley.edu (Dan Strick) of "Mon, 20 Apr 1998 00:17:02 PDT." <199804200717.AAA02451@math.berkeley.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 18:39:31 -0500 Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Dan Strick writes: > It ought to work by default. If not, you may be able to tell the > host adapter via its BIOS interface to only do narrow transfers > with specific SCSI devices. That's sorta hard to do with a narrow controller as all I've seen are not capable of wide negotiations therefore its not an option. My 2940 (narrow with the 7860 chipset and BIOS 1.21) didn't like a wide drive that was attached. -- David Kelly N4HHE, dkelly@nospam.hiwaay.net ===================================================================== The human mind ordinarily operates at only ten percent of its capacity -- the rest is overhead for the operating system. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 16:51:30 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA00298 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 16:51:30 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from ns.mt.sri.com (sri-gw.MT.net [206.127.105.141]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA00129 for ; Mon, 20 Apr 1998 23:50:34 GMT (envelope-from nate@mt.sri.com) Received: from mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.8/8.8.8) with SMTP id RAA01599; Mon, 20 Apr 1998 17:50:30 -0600 (MDT) (envelope-from nate@rocky.mt.sri.com) Received: by mt.sri.com (SMI-8.6/SMI-SVR4) id RAA03427; Mon, 20 Apr 1998 17:50:28 -0600 Date: Mon, 20 Apr 1998 17:50:28 -0600 Message-Id: <199804202350.RAA03427@mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: richard@pegasus.com (Richard Foulk) Cc: hardware@FreeBSD.ORG Subject: Re: PAO w/SSFDC pccard? In-Reply-To: <199804202100.LAA25595@pegasus.com> References: <199804202100.LAA25595@pegasus.com> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Aloha, I asked recently about connecting a PC-Card interface to an SSFDC > (SmartMedia eeprom card, for Olympus digital camera) and someone said > check out PAO ... > > Yikes! > > Okay, there's no entry in pccard.conf.sample that exactly matches my > card (several are close). But even if there was I don't understand > what the FAQ is talking about (in the Flash ATA card section) when it > says: > > (this example assumes the Flash ATA disk controller was attached > as "wdc1", and the Flash ATA disk is attached as "wd1", as > defined in the PAO config file and pccard.conf.sample) > > How is all this attaching done? With `pccardc enabler' perhaps? How? pccardd. > Before I jump in and start reverse-engineering the code, perhaps someone > can point me at some documentation I missed or provide the necessary magic. % more /etc/rc.conf % man pccardc % more /etc/pccard.conf % man pccardd Nate > crd: ctlr(0) Cirrus Logic PD672X (5 mem & 2 I/O windows) > pcic: controller I/O address 0x3e0 > pcic: controller irq 5 (shared) > pcic: controller I/O address 0x3e0 Cool. > And a few peeks with pccardc shows: > > ms# pccardc dumpcis > 2 slots found You need to update your code again. You must have Sup'd in the middle of when I broke pccardc dumpcis. :( Nate To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 17:43:47 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA09796 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 17:43:47 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from aaka.3skel.com (aaka.3skel.com [207.240.212.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id AAA09600; Tue, 21 Apr 1998 00:42:45 GMT (envelope-from danj@3skel.com) Received: from fnur.3skel.com (fnur.3skel.com [192.168.0.8]) by aaka.3skel.com (8.8.5/8.8.2) with ESMTP id UAA21813; Mon, 20 Apr 1998 20:12:10 -0400 (EDT) Received: (from danj@localhost) by fnur.3skel.com (8.8.5/8.8.2) id UAA20771; Mon, 20 Apr 1998 20:12:09 -0400 (EDT) Date: Mon, 20 Apr 1998 20:12:09 -0400 (EDT) From: Dan Janowski Message-Id: <199804210012.UAA20771@fnur.3skel.com> To: freebsd-hardware@FreeBSD.ORG, freebsd-ports@FreeBSD.ORG Subject: [Q] Sound loopthrough / Speak-freely use Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Two things that I am sure are related: I have a sound blaster 16. Sound from the microphone loops right back to the speakers. How do I turn this off? I am trying to use speak-freely and I cannot send audio and receive it at the same time. If I stop sending audio, by simply pausing sfmike, I can then hear the sound coming in off the network. Any one know why? I'm not on these mail lists. Thanks, Dan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 18:24:17 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id SAA19387 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 18:24:17 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from math.berkeley.edu (math.Berkeley.EDU [128.32.183.94]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id BAA19344 for ; Tue, 21 Apr 1998 01:24:04 GMT (envelope-from dan@math.berkeley.edu) Received: (from dan@localhost) by math.berkeley.edu (8.8.7/8.8.7) id SAA07969; Mon, 20 Apr 1998 18:24:02 -0700 (PDT) Date: Mon, 20 Apr 1998 18:24:02 -0700 (PDT) From: dan@math.berkeley.edu (Dan Strick) Message-Id: <199804210124.SAA07969@math.berkeley.edu> To: dkelly@hiwaay.net Subject: Re: Narrow SCSI controllers, and using WIDE drives with them Cc: dan@math.berkeley.edu, hardware@FreeBSD.ORG, shocking@prth.pgs.com Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > It ought to work by default. If not, you may be able to tell the > > host adapter via its BIOS interface to only do narrow transfers > > with specific SCSI devices. > > That's sorta hard to do with a narrow controller as all I've seen are > not capable of wide negotiations therefore its not an option. In that case (narrow host adapter), narrow transfers are automatic. You shouldn't have to set any options because the narrow host adapter has not got the concept of wide and should not generate the SCSI message that negotiates wide SCSI transfers. If the drive sends the message, the host adapter should reject it as an unknown message. Either way, the transfers should default to narrow. > My 2940 (narrow with the 7860 chipset and BIOS 1.21) didn't like a wide > drive that was attached. I dunno. It looks like somebody is not adhering to the standard. If you care enough to figure out who to blame, buy a SCSI bus analyzer, a copy of the SCSI-2 standard, and set aside a lot of time for playing with your new toys. It used to be that a lot of the SCSI draft standards could be found at . You can get copies of the final "published" standards only from ANSI (or in some cases from Global Engineering Documents). The SCSI-2 standard consists of one main document and several special purpose afterthoughts. The SCSI-3 standard is fragmented into a couple of dozen separate standards, most of them not yet published. I am not sure exactly which standards you will need. Wide transfers are defined in the main SCSI-2 standard, but the single 68 conductor interface is not. I know I have seen the pinout somewhere... Try the "EPI" technical report. It might tell you that I am completely wrong about this configuration working. Dan Strick dan@math.berkeley.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 19:36:26 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA00785 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 19:36:26 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from quokka.prth.tensor.pgs.com (quokka1.prth.tensor.pgs.com [157.147.224.10]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id CAA00522 for ; Tue, 21 Apr 1998 02:34:19 GMT (envelope-from shocking@ariadne.prth.tensor.pgs.com) Received: from ariadne.tensor.pgs.com (ariadne [157.147.227.36]) by quokka.prth.tensor.pgs.com (8.8.7/8.8.7) with SMTP id KAA29737; Tue, 21 Apr 1998 10:33:26 +0800 (WST) Received: from ariadne by ariadne.tensor.pgs.com (SMI-8.6/SMI-SVR4) id KAA02958; Tue, 21 Apr 1998 10:33:16 +0800 Message-Id: <199804210233.KAA02958@ariadne.tensor.pgs.com> X-Mailer: exmh version 2.0.2 2/24/98 To: David Kelly cc: hardware@FreeBSD.ORG Subject: Re: Narrow SCSI controllers, and using WIDE drives with them In-reply-to: Your message of "Mon, 20 Apr 1998 18:39:31 EST." <199804202339.SAA04391@nospam.hiwaay.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 21 Apr 1998 10:33:16 +0800 From: Stephen Hocking-Senior Programmer PGS Tensor Perth Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Dan Strick writes: > > It ought to work by default. If not, you may be able to tell the > > host adapter via its BIOS interface to only do narrow transfers > > with specific SCSI devices. > > That's sorta hard to do with a narrow controller as all I've seen are > not capable of wide negotiations therefore its not an option. > > My 2940 (narrow with the 7860 chipset and BIOS 1.21) didn't like a wide > drive that was attached. > Having started this debate, I'd like to say that I'm buying an Ultra-Wide controller (NCR-875) to go with the NCR-810 which has the narrow devices on it. I've been offered the use of an IBM 4.3GB UW drive from work. I have to boot out a PCI ethernet card, but as the box in question isn't going to be hooked up to a home 100Mbit network in the immediate future, it's not a problem. From what people have been saying, it seems like a lot of grief is involved in mxing wide & narrow devices on the same bus. Stephen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 21:16:35 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA19220 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 21:16:35 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from fly.HiWAAY.net (root@fly.HiWAAY.net [208.147.154.56]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id EAA19126 for ; Tue, 21 Apr 1998 04:16:18 GMT (envelope-from dkelly@nospam.hiwaay.net) Received: from nospam.hiwaay.net (tnt2-126.HiWAAY.net [208.147.148.126]) by fly.HiWAAY.net (8.8.8/8.8.6) with ESMTP id XAA01301; Mon, 20 Apr 1998 23:16:08 -0500 (CDT) Received: from localhost (localhost [127.0.0.1]) by nospam.hiwaay.net (8.8.8/8.8.4) with ESMTP id UAA04700; Mon, 20 Apr 1998 20:06:58 -0500 (CDT) Message-Id: <199804210106.UAA04700@nospam.hiwaay.net> X-Mailer: exmh version 2.0.2 2/24/98 To: Karl Pielorz , freebsd-hardware@FreeBSD.ORG From: David Kelly Subject: Re: Recomended ISA SCSI Cards In-reply-to: Message from "Gregory P. Smith" of "Mon, 20 Apr 1998 10:26:26 PDT." <199804201726.KAA01852@ryouko.nas.nasa.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 20:06:58 -0500 Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org "Gregory P. Smith" writes: > > Real recommendation: get a PCI motherboard, they're cheap. Seconded. A new Adaptec 1542 will demand about $125, used at $50 to $75. A new Symbios '810 PCI SCSI card is about $60. If your MB is old enough that it only has ISA then purchasing another ISA SCSI card is throwing good money after bad. Even if its only to drive a single Zip drive. Quality Asus MB's can be had for $140. Haven't been shopping so I only recal a price I saw at a recent "computer show". There was no mention as to what other SCSI devices were on the MB? Years ago I had a 486/33 that started acting funny with NE2100 ethernet cards (bus master, and picky at that). Put a cheap NE2000 in it when I added a 1542 as the NE2100 wouldn't tolerate another bus master. About 6 months later had to pull the 1542 also because the MB simply would no longer do bus mastering reliably. Initially thought it was my HD's, but they worked perfectly in another machine with the same 1542. Tried 3 different 1542's, all worked in other systems before and after the experiment. MB has been happily crunching along the past couple of years. Happily as long as it doesn't have a bus master ISA card in it. So while you are swearing at the 1542, it might not be the 1542 at all. -- David Kelly N4HHE, dkelly@nospam.hiwaay.net ===================================================================== The human mind ordinarily operates at only ten percent of its capacity -- the rest is overhead for the operating system. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Mon Apr 20 23:53:33 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA10255 for freebsd-hardware-outgoing; Mon, 20 Apr 1998 23:53:33 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from sos.freebsd.dk (sos.freebsd.dk [195.8.129.33]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id GAA10244 for ; Tue, 21 Apr 1998 06:53:25 GMT (envelope-from sos@sos.freebsd.dk) Received: (from sos@localhost) by sos.freebsd.dk (8.8.8/8.8.8) id IAA01303; Tue, 21 Apr 1998 08:53:07 +0200 (MEST) (envelope-from sos) Message-Id: <199804210653.IAA01303@sos.freebsd.dk> Subject: Re: best wdc0 flags ? In-Reply-To: <353BC450.D30240AB@tdx.co.uk> from Karl Pielorz at "Apr 20, 98 10:55:28 pm" To: kpielorz@tdx.co.uk (Karl Pielorz) Date: Tue, 21 Apr 1998 08:53:06 +0200 (MEST) Cc: mike@smith.net.au, hardware@FreeBSD.ORG From: Søren Schmidt Reply-to: sos@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL30 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In reply to Karl Pielorz who wrote: > Mike Smith wrote: > > > Try taking Soren's advice and cutting your BIOS back to mode3 PIO. He > > has actual experience with these drives, so I'd be inclined to take his > > word for it. > > I thought the BIOS config was overwritten by the wd device as it came up? - > i.e. setup every time? - In fact I thought a lot of the BIOS stuff was just > ignored anyway - and rediscovered at probe time or overwritten by the Kernel > (e.g. floppy drive type etc.) Hmm, not really, the timing selected stays.. > > Of course I do have one complicating factor... None of the drives appear in > the BIOS - if I set them in the BIOS to anything else other than 'Not > installed' the system won't boot off my SCSI drive (sd0) - the IDE's were an > afterthought (and the BIOS doesn't have a 'Boot from SCSI' option :-( Give the sucker a new BIOS :), all respectable manufactures has an upgrade for this... > If DMA is avoiding all the problems I'm probably going to leave it be... I > have another machine to setup soon - pure IDE - and suprise, suprise - it's > also getting a 4.3Gb SE - I'll try out the different BIOS options on that... > (It's also 440FX)... Nope, it will get even WORSE with DMA, you will get serious disk corruption if the timing is too fast, been there tried that, junked those Quantums :( -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Søren Schmidt (sos@FreeBSD.org) FreeBSD Core Team Even more code to hack -- will it ever end .. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Tue Apr 21 00:29:26 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA14288 for freebsd-hardware-outgoing; Tue, 21 Apr 1998 00:29:26 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from mrelay.jrc.it (mrelay.jrc.it [139.191.1.65]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA14231; Tue, 21 Apr 1998 07:29:08 GMT (envelope-from nick.hibma@jrc.it) Received: from elect8 (elect8.jrc.it [139.191.71.152]) by mrelay.jrc.it (LMC5688) with SMTP id JAA09833; Tue, 21 Apr 1998 09:28:37 +0200 (MET DST) Date: Tue, 21 Apr 1998 09:25:42 +0200 (MET DST) From: Nick Hibma X-Sender: n_hibma@elect8 Reply-To: Nick Hibma To: Mike Smith cc: "Gregory P. Smith" , freebsd-hackers@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG, Mark Willey Subject: Re: FreeBSD USB project, help requested In-Reply-To: <199804201907.MAA01295@dingo.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I'm not sure I follow you here. You can locate [OU]HCI devices by their > PCI class/subclass codes without needing to resort to vendor/product > IDs. > > > The final result should be according to [OU]HCI spec and chipset > > independent, I agree, but for the moment we do not have the knowledge > > nor the base to claim that it is chipset independent. > > If we cleave to the standard to start with, we'll be better off than > trying to separate the standard-compliant parts from the > chipset-specific parts later on... We are talking about the same thing, I think. It's just that I do not want to put the label 'UHCI compliant' on it when I have tested it with only one chip. Second, you need to recognise the chip to be able to apply the quirks: PIIX4: Switch on interrupt, supports power protection, etc. Three, from the code you would be able to derive that PIIX4 is only mentioned in the uhc_pci.c code to print a nice string at boot time. Four, I need alpha testers to see if, working from the UHCI spec and not the chip spec (I only found out last sunday that the PIIX4 supports power protection), I created something which is UHCI compliant. It is just too early to claim that. I am not working at a marketing company like MicroSchoft. Nick STA-ISIS, Joint Research Centre, Italy building: 27A tel.: +39 332 78 9549 fax.: +39 332 78 9185 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Tue Apr 21 08:38:47 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id IAA26519 for freebsd-hardware-outgoing; Tue, 21 Apr 1998 08:38:47 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from gibson.bathl.westnet.net.uk (cyberc.demon.co.uk [193.237.161.19]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA26503 for ; Tue, 21 Apr 1998 15:38:34 GMT (envelope-from graham@cyberc.demon.co.uk) Received: (from graham@localhost) by gibson.bathl.westnet.net.uk (8.8.7/8.6.9) id QAA04050 for hardware@freebsd.org; Tue, 21 Apr 1998 16:39:50 +0100 (BST) From: Graham Breach Message-Id: <199804211539.QAA04050@gibson.bathl.westnet.net.uk> Subject: Smarty (again) To: hardware@FreeBSD.ORG Date: Tue, 21 Apr 1998 16:39:50 +0100 (BST) X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I recently posted that I am trying to get the Smarty SmartDiskette to work... I think I have identified the problem - the information is stored on sectors 1, 100 and 200, but there are no sectors in between (actually, there is no rotating surface either, so "in between" is a bit stupid). Using fdcontrol I can set the number of sectors to 200, and dd will attempt to read 512 bytes from wherever, but the driver always tries to read 4 sectors at a time. Can anyone point me to the source file where I can change this value (actually bp->b_bcount for fdstrategy() ). Any help would be appreciated. Graham Breach To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Tue Apr 21 09:39:42 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA09274 for freebsd-hardware-outgoing; Tue, 21 Apr 1998 09:39:42 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA09141 for ; Tue, 21 Apr 1998 16:38:43 GMT (envelope-from bde@godzilla.zeta.org.au) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.7/8.8.7) id CAA04308; Wed, 22 Apr 1998 02:21:27 +1000 Date: Wed, 22 Apr 1998 02:21:27 +1000 From: Bruce Evans Message-Id: <199804211621.CAA04308@godzilla.zeta.org.au> To: graham@cyberc.demon.co.uk, hardware@FreeBSD.ORG Subject: Re: Smarty (again) Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >I think I have identified the problem - the information is stored on >sectors 1, 100 and 200, but there are no sectors in between (actually, >there is no rotating surface either, so "in between" is a bit stupid). > >Using fdcontrol I can set the number of sectors to 200, and dd will >attempt to read 512 bytes from wherever, but the driver always tries to >read 4 sectors at a time. 4 is probably for the usual buffered device braindamage (BLKDEV_IOSIZE / DEV_BSIZE = 4). >Can anyone point me to the source file where I can change this value >(actually bp->b_bcount for fdstrategy() ). Just use the raw device so that it reads the amount requested. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Tue Apr 21 12:16:25 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA24553 for freebsd-hardware-outgoing; Tue, 21 Apr 1998 12:16:25 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA24533 for ; Tue, 21 Apr 1998 19:16:18 GMT (envelope-from gibbs@narnia.plutotech.com) Received: (from gibbs@localhost) by narnia.plutotech.com (8.8.8/8.7.3) id NAA29458; Tue, 21 Apr 1998 13:12:16 -0600 (MDT) Date: Tue, 21 Apr 1998 13:12:16 -0600 (MDT) From: "Justin T. Gibbs" Message-Id: <199804211912.NAA29458@narnia.plutotech.com> To: "Gregory P. Smith" cc: hardware@FreeBSD.ORG Subject: Re: Recomended ISA SCSI Cards Newsgroups: pluto.freebsd.hardware In-Reply-To: <199804201726.KAA01852@ryouko.nas.nasa.gov> User-Agent: tin/pre-1.4-971204 (UNIX) (FreeBSD/3.0-CURRENT (i386)) Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > My ISA recommendation is the 1542, followed by a 1522 if you don't like > the 16mb thing (really, these are your only sane choices). I would buy an AdvanSys ISA card over a 1542 as it has tagged queuing support. You need to run CAM in order to use it though. -- Justin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Tue Apr 21 18:58:59 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id SAA16521 for freebsd-hardware-outgoing; Tue, 21 Apr 1998 18:58:59 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from fly.HiWAAY.net (fly.HiWAAY.net [208.147.154.56]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id BAA16408 for ; Wed, 22 Apr 1998 01:58:23 GMT (envelope-from dkelly@nospam.hiwaay.net) Received: from nospam.hiwaay.net (tnt2-137.HiWAAY.net [208.147.148.137]) by fly.HiWAAY.net (8.8.8/8.8.6) with ESMTP id UAA09255; Tue, 21 Apr 1998 20:57:50 -0500 (CDT) Received: from localhost (localhost [127.0.0.1]) by nospam.hiwaay.net (8.8.8/8.8.4) with ESMTP id UAA18405; Tue, 21 Apr 1998 20:57:48 -0500 (CDT) Message-Id: <199804220157.UAA18405@nospam.hiwaay.net> X-Mailer: exmh version 2.0.2 2/24/98 To: Stephen Hocking-Senior Programmer PGS Tensor Perth cc: hardware@FreeBSD.ORG From: David Kelly Subject: Re: Narrow SCSI controllers, and using WIDE drives with them In-reply-to: Message from Stephen Hocking-Senior Programmer PGS Tensor Perth of "Tue, 21 Apr 1998 10:33:16 +0800." <199804210233.KAA02958@ariadne.tensor.pgs.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 21 Apr 1998 20:57:48 -0500 Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Stephen Hocking-Senior Programmer PGS Tensor Perth writes: > > Dan Strick writes: > > > It ought to work by default. If not, you may be able to tell the > > > host adapter via its BIOS interface to only do narrow transfers > > > with specific SCSI devices. > > > > That's sorta hard to do with a narrow controller as all I've seen are > > not capable of wide negotiations therefore its not an option. > > > > My 2940 (narrow with the 7860 chipset and BIOS 1.21) didn't like a wide > > drive that was attached. > > > > Having started this debate, I'd like to say that I'm buying an Ultra-Wide > controller (NCR-875) to go with the NCR-810 which has the narrow devices on > it. I've been offered the use of an IBM 4.3GB UW drive from work. I have to > boot out a PCI ethernet card, but as the box in question isn't going to be > hooked up to a home 100Mbit network in the immediate future, it's not a > problem. From what people have been saying, it seems like a lot of grief is > involved in mxing wide & narrow devices on the same bus. While I have not been able to put wide devices on a narrow controller, I have had no trouble putting narrow devices on a wide controller. Favorite way to do it is to use a wide internal cable all the way, with a wide device at the end and its internal termination enabled. Then put the narrow devices in the middle with $13 adapters from http://www.asacomputers.com/ (and certianly elsewhere). What works almost as good is to use both the narrow and wide internal connectors on the Asus SC875. But then you can't use the external connector ("Thou Shall Not Make A 'T' Of Thine SCSI Bus"). -- David Kelly N4HHE, dkelly@nospam.hiwaay.net ===================================================================== The human mind ordinarily operates at only ten percent of its capacity -- the rest is overhead for the operating system. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Tue Apr 21 23:24:52 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA27034 for freebsd-hardware-outgoing; Tue, 21 Apr 1998 23:24:52 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from our.domaintje.com (our.domaintje.com [194.178.252.9]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id GAA26977 for ; Wed, 22 Apr 1998 06:24:42 GMT (envelope-from frank@our.domaintje.com) Received: from frank@localhost by our.domaintje.com id <7786-182>; Wed, 22 Apr 1998 08:25:11 +0200 Message-ID: <19980422082510.47497@domaintje.com> Date: Wed, 22 Apr 1998 08:25:10 +0200 From: Frank Ederveen To: freebsd-hardware@FreeBSD.ORG Subject: swap-pager: I/O error, could this really be hardware failure? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org One of our servers crashed last night, now it is spitting out lots of this on the console, send brk won't help me get into the debugger put the macgine is still pingable. Someone is going over now to hook it up to a remote powerswitch: spec_getpages: I/O read error vm_fault: pager input (probably hardware) error, PID 1 failure swap_pager: I/O error - pagein failed; blkno 4232, size 8192, error 5 vm_fault: pager input (probably hardware) error, PID 3147 failure vm_fault: pager input (probably hardware) error, PID 17830 failure spec_getpages: I/O read error vm_fault: pager input (probably hardware) error, PID 1 failure .... The machine is a Ppro-200, 256Mbyte running 2.2.6-stable, March 31st. Should I replace the memory? I seem to recall that this happened once before, a few months ago. It's an Asus mainboard with 440FX chipset. I am not sure about parity. Regards, Frank Ederveen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Wed Apr 22 00:07:41 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA12086 for freebsd-hardware-outgoing; Wed, 22 Apr 1998 00:07:41 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from implode.root.com (implode.root.com [198.145.90.17]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA12002 for ; Wed, 22 Apr 1998 07:07:31 GMT (envelope-from root@implode.root.com) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.5/8.8.5) with ESMTP id AAA09200; Wed, 22 Apr 1998 00:05:55 -0700 (PDT) Message-Id: <199804220705.AAA09200@implode.root.com> To: Frank Ederveen cc: freebsd-hardware@FreeBSD.ORG Subject: Re: swap-pager: I/O error, could this really be hardware failure? In-reply-to: Your message of "Wed, 22 Apr 1998 08:25:10 +0200." <19980422082510.47497@domaintje.com> From: David Greenman Reply-To: dg@root.com Date: Wed, 22 Apr 1998 00:05:55 -0700 Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >One of our servers crashed last night, now it is spitting out lots >of this on the console, send brk won't help me get into the debugger >put the macgine is still pingable. Someone is going over now to hook >it up to a remote powerswitch: > >spec_getpages: I/O read error >vm_fault: pager input (probably hardware) error, PID 1 failure >swap_pager: I/O error - pagein failed; blkno 4232, size 8192, error 5 >vm_fault: pager input (probably hardware) error, PID 3147 failure >vm_fault: pager input (probably hardware) error, PID 17830 failure >spec_getpages: I/O read error >vm_fault: pager input (probably hardware) error, PID 1 failure >.... > >The machine is a Ppro-200, 256Mbyte running 2.2.6-stable, March 31st. >Should I replace the memory? No, but you might take a close look at the disk drive. -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Wed Apr 22 13:09:13 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA07101 for freebsd-hardware-outgoing; Wed, 22 Apr 1998 13:09:13 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from our.domaintje.com (our.domaintje.com [194.178.252.9]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA06986 for ; Wed, 22 Apr 1998 20:08:51 GMT (envelope-from frank@our.domaintje.com) Received: from frank@localhost by our.domaintje.com id <7744-182>; Wed, 22 Apr 1998 22:09:15 +0200 Message-ID: <19980422220913.53347@domaintje.com> Date: Wed, 22 Apr 1998 22:09:13 +0200 From: Frank Ederveen To: dg@root.com, Frank Ederveen Cc: freebsd-hardware@FreeBSD.ORG Subject: Re: swap-pager: I/O error, could this really be hardware failure? References: <19980422082510.47497@domaintje.com> <199804220705.AAA09200@implode.root.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: <199804220705.AAA09200@implode.root.com>; from David Greenman on Wed, Apr 22, 1998 at 12:05:55AM -0700 Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello David and others, On Wed, Apr 22, 1998 at 12:05:55AM -0700, David Greenman wrote: > >The machine is a Ppro-200, 256Mbyte running 2.2.6-stable, March 31st. > >Should I replace the memory? > > No, but you might take a close look at the disk drive. Hmm.. It might be time to run CAM on it then; 3940-UW with Seagate drives (barracuda and cheetah). I never found any scsi-errors or retries in any of the logs. Next time I'm near it I'll give it a verify from the adaptec bios. *ponder* If I do a verify from FreeBSD, with the scsi-command, what happens if it finds an error? Report it, ask me what to do or just map it out? Swap is on sd0: # scsi -f /dev/rsd0 -m 1 AWRE (Auto Write Reallocation Enbld): 1 ARRE (Auto Read Reallocation Enbld): 1 TB (Transfer Block): 0 RC (Read Continuous): 0 EER (Enable Early Recovery): 1 PER (Post Error): 0 DTE (Disable Transfer on Error): 0 DCR (Disable Correction): 0 Read Retry Count: 82 Correction Span: 104 Head Offset Count: 0 Data Strobe Offset Count: 0 Write Retry Count: 46 Recovery Time Limit: 65535 These values are the same for all 7 drives... FrankE To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Wed Apr 22 13:57:51 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA22767 for freebsd-hardware-outgoing; Wed, 22 Apr 1998 13:57:51 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA22721 for ; Wed, 22 Apr 1998 20:57:41 GMT (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id NAA00795; Wed, 22 Apr 1998 13:54:09 -0700 (PDT) Message-Id: <199804222054.NAA00795@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: Frank Ederveen cc: freebsd-hardware@FreeBSD.ORG Subject: Re: swap-pager: I/O error, could this really be hardware failure? In-reply-to: Your message of "Wed, 22 Apr 1998 08:25:10 +0200." <19980422082510.47497@domaintje.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 22 Apr 1998 13:54:08 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > One of our servers crashed last night, now it is spitting out lots > of this on the console, send brk won't help me get into the debugger > put the macgine is still pingable. Someone is going over now to hook > it up to a remote powerswitch: > > spec_getpages: I/O read error > vm_fault: pager input (probably hardware) error, PID 1 failure > swap_pager: I/O error - pagein failed; blkno 4232, size 8192, error 5 > vm_fault: pager input (probably hardware) error, PID 3147 failure > vm_fault: pager input (probably hardware) error, PID 17830 failure > spec_getpages: I/O read error > vm_fault: pager input (probably hardware) error, PID 1 failure > ..... > > The machine is a Ppro-200, 256Mbyte running 2.2.6-stable, March 31st. > Should I replace the memory? I seem to recall that this happened once > before, a few months ago. It's an Asus mainboard with 440FX chipset. > I am not sure about parity. There should be disk errors associated with this as well. Is the machine running binaries off an NFS-mounted filesystem? -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Wed Apr 22 22:02:13 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA03072 for freebsd-hardware-outgoing; Wed, 22 Apr 1998 22:02:13 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from comp.polyu.edu.hk (csns02.COMP.POLYU.EDU.HK [158.132.25.95]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id WAA03066 for ; Wed, 22 Apr 1998 22:02:09 -0700 (PDT) (envelope-from c5666305@comp.polyu.edu.hk) Received: from cssolar82.COMP.HKP.HK by comp.polyu.edu.hk (SMI-8.6/SMI-SVR4) id MAA16320; Thu, 23 Apr 1998 12:20:21 +0800 Received: (from c5666305@localhost) by cssolar82.COMP.HKP.HK (SMI-8.6/) id MAA10861; Thu, 23 Apr 1998 12:20:19 +0800 Message-Id: <199804230420.MAA10861@cssolar82.COMP.HKP.HK> Subject: laptop for FreeBSD To: hardware@FreeBSD.ORG Date: Thu, 23 Apr 1998 12:20:19 +0800 (HKT) From: "c5666305" Cc: c5666305@csns02.comp.polyu.edu.hk X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello, I have searched and reviewed the PAO and found that not much of the record not latop running FreeBSD under XFree86 3.3.x environment. I would like to know if anyone whose machine is running XFree86 3.3.2, please send the configuration of your machine to me as I am going to buy a laptop for my final year project presentation and for future use. My idea machine may be IBM thinkpad 535x (I didn't quite the series number). Any suggestion? Thanks. Clarence To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Wed Apr 22 22:47:52 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA08790 for freebsd-hardware-outgoing; Wed, 22 Apr 1998 22:47:52 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from golem.belabm.by (root@golem.belabm.by [194.226.122.185]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id WAA07913 for ; Wed, 22 Apr 1998 22:37:08 -0700 (PDT) (envelope-from scaner@belabm.by) Received: from belabm.by ([194.226.122.183]) by golem.belabm.by (8.8.7/8.8.4) with ESMTP id IAA13745; Thu, 23 Apr 1998 08:20:49 +0300 Message-ID: <353ECF59.5BD1471E@belabm.by> Date: Thu, 23 Apr 1998 08:19:21 +0300 From: Eugene Vedistchev Reply-To: scaner@belabm.by Organization: Global One in Belarus X-Mailer: Mozilla 4.05 [en] (Win95; I) MIME-Version: 1.0 To: Peter Jeremy CC: freebsd-hardware@FreeBSD.ORG Subject: Re: Compaq Deskpro 6000 SCSI controller References: <199804202146.HAA29589@gsms01.alcatel.com.au> Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Peter Jeremy wrote: > > For reasons beyond my control, I may be forced to get a Deskpro 6000. > Does anyone know if the built-in SCSI controller is supported by > FreeBSD? AFAIK - Symbios Logic (NCR) 875 > I know the integrated NIC isn't supported. Are there any other gotcha's > that I should be aware of? > > Peter > -- > Peter Jeremy (VK2PJ) peter.jeremy@alcatel.com.au > Alcatel Australia Limited > 41 Mandible St Phone: +61 2 9690 5019 > ALEXANDRIA NSW 2015 Fax: +61 2 9690 5247 > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-hardware" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Thu Apr 23 07:33:40 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA06365 for freebsd-hardware-outgoing; Thu, 23 Apr 1998 07:33:40 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from kaiwan.kaiwan.com (kaiwan.kaiwan.com [198.178.203.2]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA06319 for ; Thu, 23 Apr 1998 07:33:18 -0700 (PDT) (envelope-from frank@exit.com) Received: from exit.com (uucp@localhost) by kaiwan.kaiwan.com (8.8.8/8.8.8) with UUCP id HAA04994; Thu, 23 Apr 1998 07:30:48 -0700 (PDT) Received: (from frank@localhost) by exit.com (8.8.8/8.7.3) id HAA05234; Thu, 23 Apr 1998 07:22:58 -0700 (PDT) From: Frank Mayhar Message-Id: <199804231422.HAA05234@exit.com> Subject: Re: Compaq Deskpro 6000 SCSI controller To: scaner@belabm.by Date: Thu, 23 Apr 1998 07:22:57 -0700 (PDT) Cc: Peter.Jeremy@alcatel.com.au, freebsd-hardware@FreeBSD.ORG In-Reply-To: <353ECF59.5BD1471E@belabm.by> from Eugene Vedistchev at "Apr 23, 98 08:19:21 am" X-Mailer: ELM [version 2.4ME+ PL17 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Eugene Vedistchev wrote: > > I know the integrated NIC isn't supported. Are there any other gotcha's > > that I should be aware of? The NIC is (theoretically) TLAN-based, and there are specs for those at http://www-s.ti.com/cgi-bin/sc/device1.cgi?device=TNETE100A&return=NETWORKING+COMPONENTS&des=THUNDERLAN++PCI+FAST+ETHERNET+CONTROLLER&type=2 (Sorry for the wrapped line.) You can get a copy of the programmer's ref there in PDF format. I have a copy, and a 6000 with one of these in it, but very little time to actually write the driver (made worse by the fact that I've never written a NIC driver before). So if someone wants to grab this information, and write a driver with it, great! OR, if someone with knowledge of these drivers can point me to documents containing tips, hints, etc., regarding writing such a driver, that would be great, too, but probably won't result in a working driver as quickly. -- Frank Mayhar frank@exit.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Thu Apr 23 09:55:06 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA08593 for freebsd-hardware-outgoing; Thu, 23 Apr 1998 09:55:06 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from kaori.communique.net (kaori.communique.net [204.27.67.55]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id JAA08389 for ; Thu, 23 Apr 1998 09:54:32 -0700 (PDT) (envelope-from rzig@verio.net) Received: by kaori.communique.net with Internet Mail Service (5.0.1458.49) id ; Thu, 23 Apr 1998 11:53:17 -0500 Message-ID: From: Raul Zighelboim To: "'hardware@freebsd.org'" Subject: ASUS P2B-S motherboard.. Date: Thu, 23 Apr 1998 11:53:14 -0500 X-Priority: 3 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.0.1458.49) Content-Type: text/plain Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello there; does any one has experience running the Asus P2B-S Motherboard with 1 Gbyte of ram ? Does it perform ? Thanks. ================================================== Raul Zighelboim rzig@verio.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Thu Apr 23 14:59:40 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA04430 for freebsd-hardware-outgoing; Thu, 23 Apr 1998 14:59:40 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from gatekeeper.alcatel.com.au (gatekeeper.alcatel.com.au [203.17.66.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id OAA04411 for ; Thu, 23 Apr 1998 14:59:25 -0700 (PDT) (envelope-from peter.jeremy@alcatel.com.au) Received: from mfg1.cim.alcatel.com.au ([139.188.23.1]) by gatekeeper.alcatel.com.au (PMDF V5.1-7 #U2695) with ESMTP id <01IW8NSL3UM8000L3O@gatekeeper.alcatel.com.au> for hardware@FreeBSD.ORG; Fri, 24 Apr 1998 07:58:48 +1000 Received: from cbd.alcatel.com.au by cim.alcatel.com.au (PMDF V5.1-10 #23324) with ESMTP id <01IW8NSHOC8WC2K7MI@cim.alcatel.com.au> for hardware@FreeBSD.ORG; Fri, 24 Apr 1998 07:58:43 +1000 Received: from gsms01.alcatel.com.au by cbd.alcatel.com.au (PMDF V5.1-7 #U2695) with ESMTP id <01IW8NSESWAOAZTTFX@cbd.alcatel.com.au> for hardware@FreeBSD.ORG; Fri, 24 Apr 1998 07:58:39 +1100 Received: (from jeremyp@localhost) by gsms01.alcatel.com.au (8.8.8/8.7.3) id HAA26070 for hardware@FreeBSD.ORG; Fri, 24 Apr 1998 07:58:37 +1000 (EST) Date: Fri, 24 Apr 1998 07:58:37 +1000 (EST) From: Peter Jeremy Subject: Re: Compaq Deskpro 6000 SCSI controller To: hardware@FreeBSD.ORG Message-id: <199804232158.HAA26070@gsms01.alcatel.com.au> Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, 23 Apr 1998 07:22:57 -0700 (PDT), Frank Mayhar wrote: >The NIC is (theoretically) TLAN-based, The NIC _is_ a TI ThunderLAN (although Compaq have changed the PCI ID codes), and the specs are readily available from TI. What's missing is a FreeBSD driver for it. If anyone is interested, there is a Linux driver for the ThunderLan chip (see ftp://ftp.caldera.com/pub/stuff/tlan-0.30.tar.gz - or maybe a later version by now) that we are using here on some Linux boxes here. I don't know how difficult it would be to port. If I do wind up with a 6000, I'll wind up porting or writing a driver (but I'd prefer not to have to do all the work and am hoping to get alternative hardware). Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 01:38:26 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA22803 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 01:38:26 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from euthyphro.uchicago.edu (euthyphro.uchicago.edu [128.135.21.31]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id BAA22795 for ; Fri, 24 Apr 1998 01:38:16 -0700 (PDT) (envelope-from sfarrell@phaedrus.uchicago.edu) Received: from phaedrus.uchicago.edu (phaedrus [128.135.21.10]) by euthyphro.uchicago.edu (8.8.6/8.8.4) with ESMTP id DAA13143 for ; Fri, 24 Apr 1998 03:37:58 -0500 (CDT) Received: (from sfarrell@localhost) by phaedrus.uchicago.edu (8.8.8/8.8.5) id DAA26673; Fri, 24 Apr 1998 03:37:52 -0500 (CDT) To: freebsd-hardware@FreeBSD.ORG Subject: 100mhz m/b's From: sfarrell+lists@farrell.org Mime-Version: 1.0 (generated by tm-edit 7.108) Content-Type: text/plain; charset=US-ASCII Date: 24 Apr 1998 03:37:52 -0500 Message-ID: <8767jzlin3.fsf@phaedrus.uchicago.edu> Lines: 18 X-Mailer: Gnus v5.6.3/XEmacs 20.4 - "Emerald" Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm thinking about getting one of these new dual pII motherboards that runs at 100mhz and holds 350 or 400mhz pII's. Anyone have any suggestions or any experience with them? I'm interested in running possible both stable and current on this box (yes, i'd run stable with one cpu only). One that looked good was: TIGER 100 DUAL PENTII INT EL440BX AGP (tyan) ($280 at necx) Thanks -- Steve Farrell To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 01:58:42 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA27068 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 01:58:42 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from heart.agmar.ru (heart.agmar.ru [195.133.66.2]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id BAA27047; Fri, 24 Apr 1998 01:58:33 -0700 (PDT) (envelope-from gvs@agmar.ru) Received: from heart.agmar.ru (heart.agmar.ru [195.133.66.2]) by heart.agmar.ru (8.8.5/8.8.5) with SMTP id MAA21963; Fri, 24 Apr 1998 12:59:47 +0400 (MSD) Date: Fri, 24 Apr 1998 12:59:47 +0400 (MSD) From: GvS One To: freebsd-hardware@FreeBSD.ORG cc: freebsd-questions@FreeBSD.ORG Subject: [Q] SuperMicro m/b models P6SBS & P6DBS Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello, folks! Did somebody ran FreeBSD 2.2.6-RELEASE or SNAP-3.0 on these motherboards? If so, what are the (possible) problems and how to solve them, or, even, what is the better choice? The second interest is m/b Iwill DPIILS2 with the same [Q]s. SY, Seva Gluschenko, just stranger at the Road. --- IRC: erra * Origin: gone to the Internet (gvs@agmar.ru) [http://www.agmar.ru/~gvs/] To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 09:03:29 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA25724 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 09:03:29 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from cerberus.partsnow.com (gatekeeper.partsnow.com [207.155.26.98] (may be forged)) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id JAA25687; Fri, 24 Apr 1998 09:03:17 -0700 (PDT) (envelope-from don@partsnow.com) Received: (from bin@localhost) by cerberus.partsnow.com (8.8.5/8.6.9) id JAA10913; Fri, 24 Apr 1998 09:02:56 -0700 (PDT) X-Authentication-Warning: cerberus.partsnow.com: bin set sender to using -f Received: from wildeweb(192.168.100.10) by cerberus.partsnow.com via smap (V2.0) id xma010906; Fri, 24 Apr 98 09:02:39 -0700 Message-ID: <3540B774.CAB68C0C@partsnow.com> Date: Fri, 24 Apr 1998 09:01:56 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net Organization: Soligen, Incorporated X-Mailer: Mozilla 4.04 [en] (X11; I; FreeBSD 2.2.5-RELEASE i386) MIME-Version: 1.0 To: freebsd-advocacy@FreeBSD.ORG CC: freebsd-hardware@FreeBSD.ORG Subject: *** Real Action Item: SPECweb Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I have spoken to several of my lower level salesman types at various companies, and they all could _immediately_ see the benefit to their companies of such a media circus as I'm proposing. (For those on -hardware, seek this thread in the -advocacy archives, I won't repeat it here) With that in mind, I'd like to solicit comments on what the best hardware (single CPU) we can get to give us the highest TCPIP network throughput. If you'll look at http://www.SPECbench.org, you'll see the disclosures for the systems Novell and others have used to achieve their performance. The key to this thing is maximum bandwidth, disk performance, and maximum intelligence on the IO channels. Even if we do it by using a gigabit card set and fiber, we're after exposure for FreeBSD and Apache, not necessarily long-lasting benchmark numbers. Once we have the system spec'd, I have no doubt I can get hardware, because it's very obvious that this will be an enormous bonanza for all major corporate sponsors. -- oooOOO O O O o * * * * * * o ___ _________ _________ ________ _________ _________ ___==_ V_=_=_DW ===--- Don Wilde [don@PartsNow.com] [http://www.PartsNow.com ] /oo0000oo-oo--oo-ooo---ooo-ooo---ooo-ooo--ooo-ooo---ooo-ooo---ooo-oo--oo  To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 09:16:33 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA27738 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 09:16:33 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from mail.webspan.net (root@mail.webspan.net [206.154.70.7]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id JAA27727; Fri, 24 Apr 1998 09:16:19 -0700 (PDT) (envelope-from opsys@mail.webspan.net) Received: from orion.webspan.net (orion.webspan.net [206.154.70.5]) by mail.webspan.net (WEBSPAN/970608) with SMTP id MAA13993; Fri, 24 Apr 1998 12:11:37 -0400 (EDT) Date: Fri, 24 Apr 1998 12:16:17 -0400 (EDT) From: Open Systems Networking X-Sender: opsys@orion.webspan.net To: dwilde1@ibm.net cc: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <3540B774.CAB68C0C@partsnow.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Don Wilde wrote: > With that in mind, I'd like to solicit comments on what the best hardware > (single CPU) we can get to give us the highest TCPIP network throughput. If > you'll look at http://www.SPECbench.org, you'll see the disclosures for the > systems Novell and others have used to achieve their performance. The key to > this thing is maximum bandwidth, disk performance, and maximum intelligence on > the IO channels. Even if we do it by using a gigabit card set and fiber, we're > after exposure for FreeBSD and Apache, not necessarily long-lasting benchmark > numbers. I have no doubt in my mind that the intel card david is using in WCARCHIVE is THE card to pump out data in insane quantities. I used to love the DEC based cards and still do, and the new tx0 SMC driver is nice to, but I think the Intel Pro/100B witht he fxp driver is THE card to use. get a MB with 4 PCI slots and slam 4 of those bad boys in there. I dont know if there are multiport versions of them but even if not these cards spit fire. Thats for PCI if you want anything else im not sure. I dont know about the ATM driver never used it. We should see what david says. I think even just 4 single port 100B's could outdo by a nice margin whatever novell used. IMO anyway. David? Generally TYAN MB's I think are the best. But it all depends on the CPU/Chipset combo. So i'm going to go look at the spechweb benchmark place before hunting down good combo's. Chris -- "I am closed minded. It keeps the rain out." ===================================| Open Systems Networking And Consulting. FreeBSD 2.2.6 is available now! | Phone: 316-326-6800 -----------------------------------| 1402 N. Washington, Wellington, KS-67152 FreeBSD: The power to serve! | E-Mail: opsys@open-systems.net http://www.freebsd.org | Consulting-Network Engineering-Security ===================================| http://open-systems.net -----BEGIN PGP PUBLIC KEY BLOCK----- Version: 2.6.2 mQENAzPemUsAAAEH/06iF0BU8pMtdLJrxp/lLk3vg9QJCHajsd25gYtR8X1Px1Te gWU0C4EwMh4seDIgK9bzFmjjlZOEgS9zEgia28xDgeluQjuuMyUFJ58MzRlC2ONC foYIZsFyIqdjEOCBdfhH5bmgB5/+L5bjDK6lNdqD8OAhtC4Xnc1UxAKq3oUgVD/Z d5UJXU2xm+f08WwGZIUcbGcaonRC/6Z/5o8YpLVBpcFeLtKW5WwGhEMxl9WDZ3Kb NZH6bx15WiB2Q/gZQib3ZXhe1xEgRP+p6BnvF364I/To9kMduHpJKU97PH3dU7Mv CXk2NG3rtOgLTEwLyvtBPqLnbx35E0JnZc0k5YkABRO0JU9wZW4gU3lzdGVtcyA8 b3BzeXNAb3Blbi1zeXN0ZW1zLm5ldD4= =BBjp -----END PGP PUBLIC KEY BLOCK----- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 10:21:17 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA06752 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 10:21:17 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from bsd.synx.com (rt.synx.com [194.167.81.239]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id KAA06641; Fri, 24 Apr 1998 10:20:54 -0700 (PDT) (envelope-from <@rn.synx.com:root@synx.com>) Received: from s3.synx.com (s3 [192.1.1.247]) by bsd.synx.com (8.6.12/8.6.12) with SMTP id SAA01817; Fri, 24 Apr 1998 18:23:41 +0100 Received: from rn by s3.synx.com id aa27782; 24 Apr 98 19:08 BST Date: Fri, 24 Apr 1998 19:18:38 +0200 (CEST) From: Remy NONNENMACHER Reply-To: remy@synx.com Subject: Re: *** Real Action Item: SPECweb To: dwilde1@ibm.net cc: don@partsnow.com, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG In-Reply-To: <3540B774.CAB68C0C@partsnow.com> MIME-Version: 1.0 Content-Type: TEXT/plain; CHARSET=US-ASCII Message-ID: <9804241908.aa27782@s3.synx.com> Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On 24 Apr, Don Wilde wrote: > I have spoken to several of my lower level salesman types at various companies, > and they all could _immediately_ see the benefit to their companies of such a > media circus as I'm proposing. (For those on -hardware, seek this thread in the > -advocacy archives, I won't repeat it here) > > With that in mind, I'd like to solicit comments on what the best hardware > (single CPU) we can get to give us the highest TCPIP network throughput. If > you'll look at http://www.SPECbench.org, you'll see the disclosures for the > systems Novell and others have used to achieve their performance. The key to > this thing is maximum bandwidth, disk performance, and maximum intelligence on > the IO channels. Even if we do it by using a gigabit card set and fiber, we're > after exposure for FreeBSD and Apache, not necessarily long-lasting benchmark > numbers. > > Once we have the system spec'd, I have no doubt I can get hardware, because it's > very obvious that this will be an enormous bonanza for all major corporate > sponsors. Interesting points about the specs : "Cache volume: NetWare volume spanning six drives, each with a single 204MB partition" This means they used stripping and only the beginning of disks (where the media rate is the highest). Hard to beat without a well tuned ccd and less than 2 SCSI controllers. (note they also used 10K rpm disks). (PS : 6x4.3Gig for creating a 1.2G filesystem !! what an efficiency !! 4% efficiency is far from a realistic system). Other parameters tend to indicate that all goes to memory and stick there. Is there any specification about a cool-down stage between tests ? Network: "All nets at half duplex". Why not full dup ? they can't ? something hidden with the BT350 ? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 10:41:51 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA09505 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 10:41:51 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from misery.sdf.com (misery.sdf.com [204.244.213.32]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id KAA09375; Fri, 24 Apr 1998 10:41:34 -0700 (PDT) (envelope-from tom@sdf.com) Received: from tom by misery.sdf.com with smtp (Exim 1.82 #3) id 0ySm4C-0002hj-00; Fri, 24 Apr 1998 10:15:16 -0700 Date: Fri, 24 Apr 1998 10:14:51 -0700 (PDT) From: Tom To: Remy NONNENMACHER cc: dwilde1@ibm.net, don@partsnow.com, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <9804241908.aa27782@s3.synx.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Remy NONNENMACHER wrote: > This means they used stripping and only the beginning of disks (where > the media rate is the highest). Hard to beat without a well tuned ccd > and less than 2 SCSI controllers. (note they also used 10K rpm disks). Better off using the hardware RAID0 in a DPT PM334UW. It also has 64MB of write-back cache so, it can re-order i/o to be more efficient. Also, using only tiny bit of each hard drive improves drive level caching. Most drives have only a 1MB cache. Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 10:43:16 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA09943 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 10:43:16 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from misery.sdf.com (misery.sdf.com [204.244.213.32]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id KAA09913; Fri, 24 Apr 1998 10:43:03 -0700 (PDT) (envelope-from tom@sdf.com) Received: from tom by misery.sdf.com with smtp (Exim 1.82 #3) id 0ySm5o-0002i1-00; Fri, 24 Apr 1998 10:16:56 -0700 Date: Fri, 24 Apr 1998 10:16:55 -0700 (PDT) From: Tom To: Open Systems Networking cc: dwilde1@ibm.net, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Open Systems Networking wrote: > think the Intel Pro/100B witht he fxp driver is THE card to use. get a MB > with 4 PCI slots and slam 4 of those bad boys in there. I dont know if That would leave no slots for a disk controller. Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 10:47:37 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA11288 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 10:47:37 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from mail.webspan.net (root@mail.webspan.net [206.154.70.7]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id KAA11125; Fri, 24 Apr 1998 10:47:01 -0700 (PDT) (envelope-from opsys@mail.webspan.net) Received: from orion.webspan.net (orion.webspan.net [206.154.70.5]) by mail.webspan.net (WEBSPAN/970608) with SMTP id NAA05306; Fri, 24 Apr 1998 13:42:01 -0400 (EDT) Date: Fri, 24 Apr 1998 13:46:41 -0400 (EDT) From: Open Systems Networking X-Sender: opsys@orion.webspan.net To: Tom cc: dwilde1@ibm.net, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Tom wrote: > > On Fri, 24 Apr 1998, Open Systems Networking wrote: > > > think the Intel Pro/100B witht he fxp driver is THE card to use. get a MB > > with 4 PCI slots and slam 4 of those bad boys in there. I dont know if > That would leave no slots for a disk controller. Unless its already built into the MB :) -- "I am closed minded. It keeps the rain out." ===================================| Open Systems Networking And Consulting. FreeBSD 2.2.6 is available now! | Phone: 316-326-6800 -----------------------------------| 1402 N. Washington, Wellington, KS-67152 FreeBSD: The power to serve! | E-Mail: opsys@open-systems.net http://www.freebsd.org | Consulting-Network Engineering-Security ===================================| http://open-systems.net -----BEGIN PGP PUBLIC KEY BLOCK----- Version: 2.6.2 mQENAzPemUsAAAEH/06iF0BU8pMtdLJrxp/lLk3vg9QJCHajsd25gYtR8X1Px1Te gWU0C4EwMh4seDIgK9bzFmjjlZOEgS9zEgia28xDgeluQjuuMyUFJ58MzRlC2ONC foYIZsFyIqdjEOCBdfhH5bmgB5/+L5bjDK6lNdqD8OAhtC4Xnc1UxAKq3oUgVD/Z d5UJXU2xm+f08WwGZIUcbGcaonRC/6Z/5o8YpLVBpcFeLtKW5WwGhEMxl9WDZ3Kb NZH6bx15WiB2Q/gZQib3ZXhe1xEgRP+p6BnvF364I/To9kMduHpJKU97PH3dU7Mv CXk2NG3rtOgLTEwLyvtBPqLnbx35E0JnZc0k5YkABRO0JU9wZW4gU3lzdGVtcyA8 b3BzeXNAb3Blbi1zeXN0ZW1zLm5ldD4= =BBjp -----END PGP PUBLIC KEY BLOCK----- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 10:49:43 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA11976 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 10:49:43 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from misery.sdf.com (misery.sdf.com [204.244.213.32]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id KAA11891; Fri, 24 Apr 1998 10:49:29 -0700 (PDT) (envelope-from tom@sdf.com) Received: from tom by misery.sdf.com with smtp (Exim 1.82 #3) id 0ySmC9-0002ii-00; Fri, 24 Apr 1998 10:23:29 -0700 Date: Fri, 24 Apr 1998 10:23:29 -0700 (PDT) From: Tom To: Open Systems Networking cc: dwilde1@ibm.net, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Open Systems Networking wrote: > On Fri, 24 Apr 1998, Tom wrote: > > > On Fri, 24 Apr 1998, Open Systems Networking wrote: > > > > > think the Intel Pro/100B witht he fxp driver is THE card to use. get a MB > > > with 4 PCI slots and slam 4 of those bad boys in there. I dont know if > > That would leave no slots for a disk controller. > > Unless its already built into the MB :) Well, you can't get a DPT PM334UW built onto any motherboards that I know of... Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 10:54:20 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA13209 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 10:54:20 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from feral.com (root@[209.54.254.2]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id KAA13176; Fri, 24 Apr 1998 10:54:06 -0700 (PDT) (envelope-from mjacob@feral.com) Received: from feral-gw (mjacob@gw100.feral.com [192.67.166.129]) by feral.com (8.8.6/8.8.6) with SMTP id KAA25378; Fri, 24 Apr 1998 10:52:10 -0700 Message-ID: <3540D149.59C4272@feral.com> Date: Fri, 24 Apr 1998 10:52:09 -0700 From: Matthew Jacob Organization: Feral Software X-Mailer: Mozilla 3.04 (X11; I; Linux 2.0.33 i586) MIME-Version: 1.0 To: Open Systems Networking CC: Tom , dwilde1@ibm.net, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I have an PR440FX (2xzPPro @ 180MHZ) motherboard. It has an onboard Pro100B and an onboard AIC78XX plus 5 PCI slots. Not bad. But, despite the bad press, the SuperMicro boards with dual PPros or Dual PII's with 8 PCI slots are still better. Tsk. It's a pity you're having to stick with Intel. An Alpha/AXP mother board (e.g., AlphaPC164) with a 600MHz processor really cranks. I mean, really. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 10:56:46 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA13850 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 10:56:46 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from mail.webspan.net (root@mail.webspan.net [206.154.70.7]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id KAA13841; Fri, 24 Apr 1998 10:56:39 -0700 (PDT) (envelope-from opsys@mail.webspan.net) Received: from orion.webspan.net (orion.webspan.net [206.154.70.5]) by mail.webspan.net (WEBSPAN/970608) with SMTP id NAA07599; Fri, 24 Apr 1998 13:51:46 -0400 (EDT) Date: Fri, 24 Apr 1998 13:56:26 -0400 (EDT) From: Open Systems Networking X-Sender: opsys@orion.webspan.net To: Matthew Jacob cc: Tom , dwilde1@ibm.net, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <3540D149.59C4272@feral.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Matthew Jacob wrote: > I have an PR440FX (2xzPPro @ 180MHZ) motherboard. It has an > onboard Pro100B and an onboard AIC78XX plus 5 PCI slots. > > Not bad. But, despite the bad press, the SuperMicro boards > with dual PPros or Dual PII's with 8 PCI slots are still better. The more PCI the merrier. > Tsk. It's a pity you're having to stick with Intel. An Alpha/AXP > mother board (e.g., AlphaPC164) with a 600MHz processor really > cranks. I mean, really. Well it isnt by CHOICE :) We don't have an alpha port yet. I still think we can pound novell's #'s 10 times over with a good selection of x86 hardware. Chris -- "I am closed minded. It keeps the rain out." ===================================| Open Systems Networking And Consulting. FreeBSD 2.2.6 is available now! | Phone: 316-326-6800 -----------------------------------| 1402 N. Washington, Wellington, KS-67152 FreeBSD: The power to serve! | E-Mail: opsys@open-systems.net http://www.freebsd..org | Consulting-Network Engineering-Security ===================================| http://open-systems.net -----BEGIN PGP PUBLIC KEY BLOCK----- Version: 2.6.2 mQENAzPemUsAAAEH/06iF0BU8pMtdLJrxp/lLk3vg9QJCHajsd25gYtR8X1Px1Te gWU0C4EwMh4seDIgK9bzFmjjlZOEgS9zEgia28xDgeluQjuuMyUFJ58MzRlC2ONC foYIZsFyIqdjEOCBdfhH5bmgB5/+L5bjDK6lNdqD8OAhtC4Xnc1UxAKq3oUgVD/Z d5UJXU2xm+f08WwGZIUcbGcaonRC/6Z/5o8YpLVBpcFeLtKW5WwGhEMxl9WDZ3Kb NZH6bx15WiB2Q/gZQib3ZXhe1xEgRP+p6BnvF364I/To9kMduHpJKU97PH3dU7Mv CXk2NG3rtOgLTEwLyvtBPqLnbx35E0JnZc0k5YkABRO0JU9wZW4gU3lzdGVtcyA8 b3BzeXNAb3Blbi1zeXN0ZW1zLm5ldD4= =BBjp -----END PGP PUBLIC KEY BLOCK----- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 10:59:07 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA14466 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 10:59:07 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id KAA14452; Fri, 24 Apr 1998 10:59:01 -0700 (PDT) (envelope-from marcs@znep.com) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.7/8.8.7) with UUCP id LAA22100; Fri, 24 Apr 1998 11:58:55 -0600 (MDT) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id LAA27836; Fri, 24 Apr 1998 11:57:54 -0600 (MDT) Date: Fri, 24 Apr 1998 11:57:53 -0600 (MDT) From: Marc Slemko To: Open Systems Networking cc: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Open Systems Networking wrote: > On Fri, 24 Apr 1998, Don Wilde wrote: > > > With that in mind, I'd like to solicit comments on what the best hardware > > (single CPU) we can get to give us the highest TCPIP network throughput. If > > you'll look at http://www.SPECbench.org, you'll see the disclosures for the > > systems Novell and others have used to achieve their performance. The key to > > this thing is maximum bandwidth, disk performance, and maximum intelligence on > > the IO channels. Even if we do it by using a gigabit card set and fiber, we're > > after exposure for FreeBSD and Apache, not necessarily long-lasting benchmark > > numbers. Whatever you do, your results had better be limited by something other than the network. The drivers are important, the card design is important, etc. but whatever you do the point of the benchmark is to not be network limited. > > I have no doubt in my mind that the intel card david is using in WCARCHIVE > is THE card to pump out data in insane quantities. I used to love the DEC > based cards and still do, and the new tx0 SMC driver is nice to, but I > think the Intel Pro/100B witht he fxp driver is THE card to use. get a MB > with 4 PCI slots and slam 4 of those bad boys in there. I dont know if > there are multiport versions of them but even if not these cards spit > fire. Thats for PCI if you want anything else im not sure. I dont know > about the ATM driver never used it. We should see what david says. I think > even just 4 single port 100B's could outdo by a nice margin whatever > novell used. IMO anyway. David? Generally TYAN MB's I think are the best. > But it all depends on the CPU/Chipset combo. > So i'm going to go look at the spechweb benchmark place before hunting > down good combo's. For web benchmarks jacking up the MTU beyond what you can get with Ethernet can give significant wins. FDDI can help a bit, ATM can help even more. Can't comment on driver support in FreeBSD, and using a NIC with a bad driver or design can hurt things a lot. See ftp://ftp.cup.hp.com/dist/networking/briefs/WebMTU.pdf for some example numbers on the impact this can have. This is the sort of thing you are dealing with when competing against SPECweb96 numbers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 11:07:32 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA16318 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 11:07:32 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA16295; Fri, 24 Apr 1998 11:07:06 -0700 (PDT) (envelope-from marcs@znep.com) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.7/8.8.7) with UUCP id MAA22418; Fri, 24 Apr 1998 12:07:02 -0600 (MDT) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id MAA28003; Fri, 24 Apr 1998 12:03:31 -0600 (MDT) Date: Fri, 24 Apr 1998 12:03:30 -0600 (MDT) From: Marc Slemko To: Open Systems Networking cc: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Open Systems Networking wrote: > Well it isnt by CHOICE :) > We don't have an alpha port yet. I still think we can pound novell's #'s > 10 times over with a good selection of x86 hardware. If you think that then you are wrong. Going into this with this sort of expectation will result in nothing but heartbreak. You may be able to come close to them if you got a good Zeus port to FreeBSD. You won't with Apache. I'm not trying to be negative, just trying to make reality crystal clear. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 11:07:54 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA16603 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 11:07:54 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA16536; Fri, 24 Apr 1998 11:07:47 -0700 (PDT) (envelope-from marcs@znep.com) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.7/8.8.7) with UUCP id MAA22426; Fri, 24 Apr 1998 12:07:17 -0600 (MDT) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id MAA28008; Fri, 24 Apr 1998 12:04:39 -0600 (MDT) Date: Fri, 24 Apr 1998 12:04:38 -0600 (MDT) From: Marc Slemko To: Tom cc: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Tom wrote: > > On Fri, 24 Apr 1998, Open Systems Networking wrote: > > > think the Intel Pro/100B witht he fxp driver is THE card to use. get a MB > > with 4 PCI slots and slam 4 of those bad boys in there. I dont know if > > That would leave no slots for a disk controller. You may want a couple of controllers, depending. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 11:08:55 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA17167 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 11:08:55 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA17155; Fri, 24 Apr 1998 11:08:47 -0700 (PDT) (envelope-from jkh@time.cdrom.com) Received: from time.cdrom.com (jkh@localhost.cdrom.com [127.0.0.1]) by time.cdrom.com (8.8.8/8.8.8) with ESMTP id LAA24858; Fri, 24 Apr 1998 11:08:35 -0700 (PDT) (envelope-from jkh@time.cdrom.com) To: Marc Slemko cc: Open Systems Networking , freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-reply-to: Your message of "Fri, 24 Apr 1998 12:03:30 MDT." Date: Fri, 24 Apr 1998 11:08:34 -0700 Message-ID: <24854.893441314@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > You may be able to come close to them if you got a good Zeus port to > FreeBSD. You won't with Apache. Uh, there is a Zeus port to FreeBSD and has been for at least 2 years no? Jordan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 11:10:22 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA17855 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 11:10:22 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA17790; Fri, 24 Apr 1998 11:10:11 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id LAA00744; Fri, 24 Apr 1998 11:07:01 -0700 (PDT) Message-Id: <199804241807.LAA00744@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: dwilde1@ibm.net cc: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-reply-to: Your message of "Fri, 24 Apr 1998 09:01:56 PDT." <3540B774.CAB68C0C@partsnow.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Apr 1998 11:07:00 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I have spoken to several of my lower level sal> Once we have the system spec'd, I have no doubt I can get hardware, because it's > very obvious that this will be an enormous bonanza for all major corporate > sponsors. - Dual 400MHz PII's (I think this will finally pull ahead of the 1M cache 233MHz P6'en). - A BX-based board with as much memory as the PII's will cache. - All of the OS, and if possible all of the web data in MFS. This may involve the biggest, nastiest PicoBSD config you can imagine. - If it won't all work in an MFS, a DPT controller and a small farm of 10000rpm fibrechannel disks. Otherwise, the disk is irrelevant. - No swap. Then talk to the various ATM and gigabit networking people that are playing with FreeBSD. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 11:16:45 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA19690 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 11:16:45 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA19676; Fri, 24 Apr 1998 11:16:32 -0700 (PDT) (envelope-from marcs@znep.com) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.7/8.8.7) with UUCP id MAA22710; Fri, 24 Apr 1998 12:15:13 -0600 (MDT) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id MAA28115; Fri, 24 Apr 1998 12:15:38 -0600 (MDT) Date: Fri, 24 Apr 1998 12:15:38 -0600 (MDT) From: Marc Slemko To: "Jordan K. Hubbard" cc: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <24854.893441314@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Jordan K. Hubbard wrote: > > You may be able to come close to them if you got a good Zeus port to > > FreeBSD. You won't with Apache. > > Uh, there is a Zeus port to FreeBSD and has been for at least 2 years > no? v1, yes. v3, not AFAIK. If you use v1 you could do it, but I don't know if Zeus implements all the stuff it needs to on FreeBSD to get the best performance. Have never used it on FreeBSD. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 11:18:24 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA20212 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 11:18:24 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from mail.webspan.net (root@mail.webspan.net [206.154.70.7]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA20146 for ; Fri, 24 Apr 1998 11:18:06 -0700 (PDT) (envelope-from opsys@mail.webspan.net) Received: from orion.webspan.net (orion.webspan.net [206.154.70.5]) by mail.webspan.net (WEBSPAN/970608) with SMTP id OAA12566; Fri, 24 Apr 1998 14:12:55 -0400 (EDT) Date: Fri, 24 Apr 1998 14:17:35 -0400 (EDT) From: Open Systems Networking X-Sender: opsys@orion.webspan.net To: "Jordan K. Hubbard" cc: Marc Slemko , freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <24854.893441314@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Jordan K. Hubbard wrote: > > You may be able to come close to them if you got a good Zeus port to > > FreeBSD. You won't with Apache. > > Uh, there is a Zeus port to FreeBSD and has been for at least 2 years > no? That was one of their first if not first platforms was FreeBSD I think. And now AHEM lets nuke the crossposting :) Chris -- "I am closed minded. It keeps the rain out." ===================================| Open Systems Networking And Consulting. FreeBSD 2.2.6 is available now! | Phone: 316-326-6800 -----------------------------------| 1402 N. Washington, Wellington, KS-67152 FreeBSD: The power to serve! | E-Mail: opsys@open-systems.net http://www.freebsd.org | Consulting-Network Engineering-Security ===================================| http://open-systems.net -----BEGIN PGP PUBLIC KEY BLOCK----- Version: 2.6.2 mQENAzPemUsAAAEH/06iF0BU8pMtdLJrxp/lLk3vg9QJCHajsd25gYtR8X1Px1Te gWU0C4EwMh4seDIgK9bzFmjjlZOEgS9zEgia28xDgeluQjuuMyUFJ58MzRlC2ONC foYIZsFyIqdjEOCBdfhH5bmgB5/+L5bjDK6lNdqD8OAhtC4Xnc1UxAKq3oUgVD/Z d5UJXU2xm+f08WwGZIUcbGcaonRC/6Z/5o8YpLVBpcFeLtKW5WwGhEMxl9WDZ3Kb NZH6bx15WiB2Q/gZQib3ZXhe1xEgRP+p6BnvF364I/To9kMduHpJKU97PH3dU7Mv CXk2NG3rtOgLTEwLyvtBPqLnbx35E0JnZc0k5YkABRO0JU9wZW4gU3lzdGVtcyA8 b3BzeXNAb3Blbi1zeXN0ZW1zLm5ldD4= =BBjp -----END PGP PUBLIC KEY BLOCK----- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 11:23:36 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA21410 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 11:23:36 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA21154; Fri, 24 Apr 1998 11:23:16 -0700 (PDT) (envelope-from marcs@znep.com) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.7/8.8.7) with UUCP id MAA22910; Fri, 24 Apr 1998 12:22:56 -0600 (MDT) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id MAA28160; Fri, 24 Apr 1998 12:23:13 -0600 (MDT) Date: Fri, 24 Apr 1998 12:23:12 -0600 (MDT) From: Marc Slemko To: Mike Smith cc: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <199804241807.LAA00744@dingo.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Mike Smith wrote: > > I have spoken to several of my lower level sal> Once we have the system spec'd, I have no doubt I can get hardware, because it's > > very obvious that this will be an enormous bonanza for all major corporate > > sponsors. > > - Dual 400MHz PII's (I think this will finally pull ahead of the 1M > cache 233MHz P6'en). I really doubt that dual processors will give you any advantage even if you were willing to use current. I'm not up on the current state of FreeBSD SMP, but if it isn't beyond the level of a single spin lock on kernel code, you won't win and you could even lose by going SMP. You can actually get worse performance when using SMP on stable Linux kernels than not using it. > - A BX-based board with as much memory as the PII's will cache. > - All of the OS, and if possible all of the web data in MFS. This may > involve the biggest, nastiest PicoBSD config you can imagine. The size of the SPECweb dataset actually changes with the "expected" benchmark results. You are probably looking at... 500-750 megs or so. I don't think you can use MFS, there are certain restrictions on what you can and can't do when reporting SPECweb numbers. > - If it won't all work in an MFS, a DPT controller and a small farm of > 10000rpm fibrechannel disks. Otherwise, the disk is irrelevant. > - No swap. No, you want swap especially if using Apache (because of all the COW pages). It won't be "really" used, but you still need it. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 11:35:53 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA25216 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 11:35:53 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA25210; Fri, 24 Apr 1998 11:35:48 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id LAA00871; Fri, 24 Apr 1998 11:31:34 -0700 (PDT) Message-Id: <199804241831.LAA00871@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: Marc Slemko cc: Mike Smith , freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-reply-to: Your message of "Fri, 24 Apr 1998 12:23:12 MDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Apr 1998 11:31:33 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > On Fri, 24 Apr 1998, Mike Smith wrote: > > > > I have spoken to several of my lower level sal> Once we have the system spec'd, I have no doubt I can get hardware, because it's > > > very obvious that this will be an enormous bonanza for all major corporate > > > sponsors. > > > > - Dual 400MHz PII's (I think this will finally pull ahead of the 1M > > cache 233MHz P6'en). > > I really doubt that dual processors will give you any advantage even if > you were willing to use current. I'm not up on the current state of > FreeBSD SMP, but if it isn't beyond the level of a single spin lock on > kernel code, you won't win and you could even lose by going SMP. You can > actually get worse performance when using SMP on stable Linux kernels than > not using it. That depends on where the load is; if it's in kernel space then I'd be inclined to agree with you (and I'd say that the 1M P6 would probably be a contender there). If the load is even substantially in user space, then the story seems to change entirely. If nothing else, having the second CPU would be educational from our perspective. > > - A BX-based board with as much memory as the PII's will cache. > > - All of the OS, and if possible all of the web data in MFS. This may > > involve the biggest, nastiest PicoBSD config you can imagine. > > The size of the SPECweb dataset actually changes with the "expected" > benchmark results. You are probably looking at... 500-750 megs or so. What sort of working-set size is the system likely to have? On a 1GB system, you could probably get away with an MFS like that. > I don't think you can use MFS, there are certain restrictions on what you > can and can't do when reporting SPECweb numbers. 8( Ok, solid-state SCSI disks then. > > - If it won't all work in an MFS, a DPT controller and a small farm of > > 10000rpm fibrechannel disks. Otherwise, the disk is irrelevant. > > - No swap. > > No, you want swap especially if using Apache (because of all the COW > pages). It won't be "really" used, but you still need it. I'm not sure I follow this, but I may be missing a nuance in our handing of COW. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 12:38:15 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA09938 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 12:38:15 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA09922; Fri, 24 Apr 1998 12:38:05 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-248.ca.us.ibm.net [32.100.79.248]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id TAA42596; Fri, 24 Apr 1998 19:37:48 GMT Message-ID: <3540E161.4149FA73@ibm.net> Date: Fri, 24 Apr 1998 12:00:49 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: Matthew Jacob CC: Open Systems Networking , Tom , freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: <3540D149.59C4272@feral.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Matthew Jacob wrote: > > Tsk. It's a pity you're having to stick with Intel. An Alpha/AXP > mother board (e.g., AlphaPC164) with a 600MHz processor really > cranks. I mean, really. Sure, and no doubt that's what the Linux guys will use against us as soon as we do this. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 12:38:20 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA10000 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 12:38:20 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA09942 for ; Fri, 24 Apr 1998 12:38:15 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-248.ca.us.ibm.net [32.100.79.248]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id TAA88256; Fri, 24 Apr 1998 19:37:58 GMT Message-ID: <3540E43F.83D3A0D5@ibm.net> Date: Fri, 24 Apr 1998 12:13:03 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: Marc Slemko CC: freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Glad you're still listening, Marc - Yep, that was the first thing I noticed. 5 _Intelligent_ 100Base-T cards in theirs. If we could find an ATM or Gigabit Ethernet card or Fibre CHannel card that would mimic a PCI net card, that would give us a boost to the next level, where the 33Mhz PCI becomes the limiting factor. If it uses standard NIC drivers, we can still stay within the guidelines of not tweaking our software beyond what's RELEASEd. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 12:38:23 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA10029 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 12:38:23 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA09910; Fri, 24 Apr 1998 12:38:03 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-248.ca.us.ibm.net [32.100.79.248]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id TAA37576; Fri, 24 Apr 1998 19:37:30 GMT Message-ID: <3540DED7.CD705B70@ibm.net> Date: Fri, 24 Apr 1998 11:49:59 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: remy@synx.com CC: don@partsnow.com, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG, brian@hyperreal.org Subject: Re: *** Real Action Item: SPECweb References: <9804241908.aa27782@s3.synx.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I saw that too. Remember, this event is going to be somewhat different than my original challenge. This is raw horsepower, and I hve no doubt that our numbers will be eclipsed as soon as somebody does an I2O system. What I want to do is to grab the world's attention long enough to show that we're a serious player, and then slam them with my original Challenge, once they can't refuse without losing face. Remember, exposure and good press is our goal! The Novell system would be a stupid system from a real world viewpoint. I wonder if there's actually something in the SPECweb benchmark that will prevent us from just loading the whole test website in a RAM disk. [Brian?] To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 12:38:28 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA10136 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 12:38:28 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA10003 for ; Fri, 24 Apr 1998 12:38:21 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-248.ca.us.ibm.net [32.100.79.248]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id TAA101650; Fri, 24 Apr 1998 19:38:09 GMT Message-ID: <3540E711.C4536E88@ibm.net> Date: Fri, 24 Apr 1998 12:25:05 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: Marc Slemko CC: "Jordan K. Hubbard" , freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I think hardware's our plus. I'd also like to stick with Apache a) because of our kind friends b) because we know it c) because it _IS_ the well known server platform and is very popular in the real world. Remember, this is all a game to get publicity for freeware, specifically us (FreeBSD and Apache). It has no relationship to the real world. What we do insofar as stretching the rules as far as possible is a tribute to our ingenuity and a plus in this instance, and I guarantee our numbers will be bested in nothing flat as soon as we tweak tails. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 12:38:42 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA10326 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 12:38:42 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA09972; Fri, 24 Apr 1998 12:38:17 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-248.ca.us.ibm.net [32.100.79.248]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id TAA24566; Fri, 24 Apr 1998 19:38:05 GMT Message-ID: <3540E505.720BE9AB@ibm.net> Date: Fri, 24 Apr 1998 12:16:21 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: Mike Smith CC: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: <199804241807.LAA00744@dingo.cdrom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org This is my thought as well, although like I said 1 CPU. Does PICOBSD support 2xCPU? BX I like, although we may have a problem getting it. We can disable all console output during the test, so that's even faster than an AGP writing text. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 12:39:13 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA10784 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 12:39:13 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA09914; Fri, 24 Apr 1998 12:38:04 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-248.ca.us.ibm.net [32.100.79.248]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id TAA83758; Fri, 24 Apr 1998 19:37:42 GMT Message-ID: <3540E0AF.C3B5C576@ibm.net> Date: Fri, 24 Apr 1998 11:57:51 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: Tom CC: Remy NONNENMACHER , don@partsnow.com, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Tom wrote: > Better off using the hardware RAID0 in a DPT PM334UW. It also has 64MB > of write-back cache so, it can re-order i/o to be more efficient. > > Also, using only tiny bit of each hard drive improves drive level > caching. Most drives have only a 1MB cache. I have a friend at Winchester Systems who's trying to sell me a SCSI-3 RAID box, and I think they'd be willing to let us borrow one for a test in return for favorable mention http://www.winsys.com. He says they work with fast differential 2944's as host cards. Is the DPT faster with the direct bus connection? No SCSI at all? Does anybody have experience with these new Compact PCI chassis that have 8 PCI slots? An obvious bottleneck for anything we can do is going to be the chassis limitation on most of 4 PCI. I believe the Compaq Proliant is EISA. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 13:04:27 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA15813 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 13:04:27 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA15706; Fri, 24 Apr 1998 13:04:13 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id NAA01264; Fri, 24 Apr 1998 13:00:30 -0700 (PDT) Message-Id: <199804242000.NAA01264@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: dwilde1@ibm.net cc: Mike Smith , freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-reply-to: Your message of "Fri, 24 Apr 1998 12:16:21 PDT." <3540E505.720BE9AB@ibm.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Apr 1998 13:00:29 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > This is my thought as well, although like I said 1 CPU. Does PICOBSD > support 2xCPU? PicoBSD is just a custom FreeBSD configuration toolkit that builds an entire system that runs out of an MFS. > BX I like, although we may have a problem getting it. We can disable all > console output during the test, so that's even faster than an AGP > writing text. Availibility for BX boards doesn't seem to be too bad. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 13:05:51 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA16381 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 13:05:51 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA16351; Fri, 24 Apr 1998 13:05:41 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id NAA01282; Fri, 24 Apr 1998 13:02:29 -0700 (PDT) Message-Id: <199804242002.NAA01282@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: dwilde1@ibm.net cc: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-reply-to: Your message of "Fri, 24 Apr 1998 11:57:51 PDT." <3540E0AF.C3B5C576@ibm.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Apr 1998 13:02:29 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Tom wrote: > > > Better off using the hardware RAID0 in a DPT PM334UW. It also has 64MB > > of write-back cache so, it can re-order i/o to be more efficient. > > > > Also, using only tiny bit of each hard drive improves drive level > > caching. Most drives have only a 1MB cache. > > I have a friend at Winchester Systems who's trying to sell me a SCSI-3 > RAID box, and I think they'd be willing to let us borrow one for a test > in return for favorable mention http://www.winsys.com. He says they work > with fast differential 2944's as host cards. Is the DPT faster with the > direct bus connection? No SCSI at all? It sounds like there's some requirement to use disks; if throughput is an issue then the DPT is probably the way to go. > Does anybody have experience with these new Compact PCI chassis that > have 8 PCI slots? An obvious bottleneck for anything we can do is going > to be the chassis limitation on most of 4 PCI. These are generally 4 slots and a bridge with 4 more behind. Depending on the configuration and the hardware in question, more slots is not necessarily better. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 13:21:57 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA21310 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 13:21:57 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from george.arc.nasa.gov (george.arc.nasa.gov [128.102.194.142]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA21062 for ; Fri, 24 Apr 1998 13:21:14 -0700 (PDT) (envelope-from lamaster@george.arc.nasa.gov) Received: (from lamaster@localhost) by george.arc.nasa.gov (8.8.7/8.8.7) id NAA04441; Fri, 24 Apr 1998 13:21:12 -0700 (PDT) Date: Fri, 24 Apr 1998 13:21:12 -0700 (PDT) From: Hugh LaMaster To: freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <3540E43F.83D3A0D5@ibm.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Don Wilde wrote: > Yep, that was the first thing I noticed. 5 _Intelligent_ 100Base-T cards > in theirs. I'm not arguing against the Intel cards, but, isn't the performance on DEC Tulip cards about as good? Any numbers somewhere on this? > If we could find an ATM or Gigabit Ethernet card or Fibre CHannel card > that would mimic a PCI net card, that would give us a boost to the next > level, where the 33Mhz PCI becomes the limiting factor. If it uses Drivers have already been written for the Packet Engines G-NIC cards. I would guess that porting the G-NIC drivers to 3.0-current should be easy. What would be on the receiving end of all this data, BTW? As for the PPro vs. 400 Mhz P II - last weekend, I checked out Intel's information and compared it to other numbers I have. Based on that, it appears that the memory bandwidth of the BX chipset w/ SDRAM is (finally) pretty good. And, it isn't on the PPro/Natoma. So, for some kind of bake-off test, I would strongly suggest getting a 400 MHz P II, a new BX-based board (e.g. SuperMicro), and 100 Mhz 8ns unbuffered "PC100 certified" SDRAM - it appears that it really will run at x-1-1-1 and hopefully get at least twice the memory bandwidth of an 233 MHz PentiumMMX on HX chipset- at least if the numbers I saw pan out. [e.g. Intel has published STREAM numbers in its Performance Brief. Looks pretty good in comparison to previous x86's.] Also, I would get the dual-processor motherboard and play around with it, but, I have to believe that the single processor version will be a lot faster in any network-bandwidth-limited test. I doubt if the bcopy()'s in the network stack are multi-threaded in the SMP kernel. MP systems are good when you have several processes with at least one in mostly user-state- even for a single user, with lots of time in the X server, and in a CPU-intensive user program, etc. But, if this test (I'm not sure exactly what test is being discussed, but, if it involves 5 100baseT cards ...) is going to be limited by the network stack, go with the single-CPU kernel/system. For the SCSI controller, it sounds like either the Adaptec 7895 or Symbios 3C876 based cards would do - the new SCSI code will support tagged queuing on either of these chipsets, correct? [The 7895 is now available built in to some motherboards, e.g., Supermicro BX boards, freeing up another slot for NICs. Has anybody got one of these boards working yet on 3.0-current? It should make a pretty decent workstation.] Now, what *Big Contest* is being discussed here anyway? -- Hugh LaMaster, M/S 233-21, ASCII Email: hlamaster@mail.arc.nasa.gov NASA Ames Research Center Or: lamaster@george.arc.nasa.gov Moffett Field, CA 94035-1000 No Junkmail: USC 18 section 2701 Phone: 650/604-1056 Disclaimer: Unofficial, personal *opinion*. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 13:22:15 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA21471 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 13:22:15 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from math.berkeley.edu (math.Berkeley.EDU [128.32.183.94]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA21341 for ; Fri, 24 Apr 1998 13:22:01 -0700 (PDT) (envelope-from dan@math.berkeley.edu) Received: (from dan@localhost) by math.berkeley.edu (8.8.7/8.8.7) id NAA13815; Fri, 24 Apr 1998 13:21:58 -0700 (PDT) Date: Fri, 24 Apr 1998 13:21:58 -0700 (PDT) From: dan@math.berkeley.edu (Dan Strick) Message-Id: <199804242021.NAA13815@math.berkeley.edu> To: freebsd-hardware@FreeBSD.ORG Subject: What is this "raid" card? Cc: dan@math.berkeley.edu Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I have recently been donated an IBM "SCSI-2 F/W PCI RAID Adapter" PCI card. Does anyone know what this is? More relevantly: is there a FreeBSD driver for it? As you might expect, it comes only with IBM RS6000/AIX drivers. So far, only Solaris-2 and a standalone configuration diskette of unknown origin recognize the device (and with Solaris-2 all that is known is that reads and writes seem to work). Dan Strick dan@math.berkeley.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 13:29:02 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA22763 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 13:29:02 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA22753 for ; Fri, 24 Apr 1998 13:28:45 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id NAA01378; Fri, 24 Apr 1998 13:25:01 -0700 (PDT) Message-Id: <199804242025.NAA01378@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: dan@math.berkeley.edu (Dan Strick) cc: freebsd-hardware@FreeBSD.ORG Subject: Re: What is this "raid" card? In-reply-to: Your message of "Fri, 24 Apr 1998 13:21:58 PDT." <199804242021.NAA13815@math.berkeley.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Apr 1998 13:25:01 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I have recently been donated an IBM "SCSI-2 F/W PCI RAID Adapter" > PCI card. Does anyone know what this is? More relevantly: is there > a FreeBSD driver for it? As you might expect, it comes only with > IBM RS6000/AIX drivers. If nobody can positively identify it, and reading the runes on the parts on the board doesn't tell you anything, you can always bring it past here for an analysis. 8) But I would start by looking at all the big chips on the card and looking for any branding other than IBM, and guessing based on that. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 13:50:22 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA26683 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 13:50:22 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA26667; Fri, 24 Apr 1998 13:50:02 -0700 (PDT) (envelope-from marcs@znep.com) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.7/8.8.7) with UUCP id OAA27391; Fri, 24 Apr 1998 14:49:45 -0600 (MDT) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id OAA28998; Fri, 24 Apr 1998 14:43:53 -0600 (MDT) Date: Fri, 24 Apr 1998 14:43:52 -0600 (MDT) From: Marc Slemko To: Mike Smith cc: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <199804241831.LAA00871@dingo.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Mike Smith wrote: > > On Fri, 24 Apr 1998, Mike Smith wrote: > > > > > > I have spoken to several of my lower level sal> Once we have the system spec'd, I have no doubt I can get hardware, because it's > > > > very obvious that this will be an enormous bonanza for all major corporate > > > > sponsors. > > > > > > - Dual 400MHz PII's (I think this will finally pull ahead of the 1M > > > cache 233MHz P6'en). > > > > I really doubt that dual processors will give you any advantage even if > > you were willing to use current. I'm not up on the current state of > > FreeBSD SMP, but if it isn't beyond the level of a single spin lock on > > kernel code, you won't win and you could even lose by going SMP. You can > > actually get worse performance when using SMP on stable Linux kernels than > > not using it. > > That depends on where the load is; if it's in kernel space then I'd be > inclined to agree with you (and I'd say that the 1M P6 would probably > be a contender there). > > If the load is even substantially in user space, then the story seems > to change entirely. If nothing else, having the second CPU would be > educational from our perspective. Webserving is very kernel heavy and I would doubt it would gain at all from SMP on a system using a single spin lock. That has been my experience on numerous operating systems. AFAIK, the issues that have to be dealt with in the SMP code to fix this are really clear (not necessarily the way to do it, but the concepts) so I don't think there would be a big benefit. > > > > - A BX-based board with as much memory as the PII's will cache. > > > - All of the OS, and if possible all of the web data in MFS. This may > > > involve the biggest, nastiest PicoBSD config you can imagine. > > > > The size of the SPECweb dataset actually changes with the "expected" > > benchmark results. You are probably looking at... 500-750 megs or so. > > What sort of working-set size is the system likely to have? On a 1GB > system, you could probably get away with an MFS like that. It depends on how much more expensive reading from a MFS file system is than reading from the buffer cache. > > > I don't think you can use MFS, there are certain restrictions on what you > > can and can't do when reporting SPECweb numbers. > > 8( Ok, solid-state SCSI disks then. There is a requirement that: The server utilizes stable storage for all data files and server logs. The log file records must be written to non-volatile storage at least as often as once per 60 seconds. A description of the rules is at http://www.spec.org/osg/web96/runrules.html > > > > - If it won't all work in an MFS, a DPT controller and a small farm of > > > 10000rpm fibrechannel disks. Otherwise, the disk is irrelevant. > > > - No swap. > > > > No, you want swap especially if using Apache (because of all the COW > > pages). It won't be "really" used, but you still need it. > > I'm not sure I follow this, but I may be missing a nuance in our > handing of COW. Swap is allocated for pages flagged COW, at the time the pages are mapped, no? If you don't have swap... Because of the design of Apache (more so on servers with more config info, thousands of vhosts, etc.) where the child processes are forked from the parent, you can end up with a lot of COW pages. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 13:53:28 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA27321 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 13:53:28 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA27302 for ; Fri, 24 Apr 1998 13:53:14 -0700 (PDT) (envelope-from marcs@znep.com) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.7/8.8.7) with UUCP id OAA27538; Fri, 24 Apr 1998 14:53:06 -0600 (MDT) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id OAA29049; Fri, 24 Apr 1998 14:52:37 -0600 (MDT) Date: Fri, 24 Apr 1998 14:52:37 -0600 (MDT) From: Marc Slemko To: Don Wilde cc: freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <3540E711.C4536E88@ibm.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Don Wilde wrote: > > I think hardware's our plus. I'd also like to stick with Apache a) > because of our kind friends b) because we know it c) because it _IS_ the > well known server platform and is very popular in the real world. > Remember, this is all a game to get publicity for freeware, specifically > us (FreeBSD and Apache). It has no relationship to the real world. What > we do insofar as stretching the rules as far as possible is a tribute to > our ingenuity and a plus in this instance, and I guarantee our numbers > will be bested in nothing flat as soon as we tweak tails. If you are trying to make a system that will beat currently listed results, you must use Zeus. I encourage the use of Apache for everything but I also know and respect its current limitations; just what you expect needs to be very clear or the whole thing will end up disappointing everyone. Apache can give good numbers and I would be interested in seeing what results you can get. They will not be as good as something like Zeus on comparable hardware. Zeus will also, generally speaking, kick ass on most memory limited systems and on CPU limited systems. Results don't have to beat everything else to prove a point and to be useful. You do, however, have to know where you are aiming. -- Marc Slemko | Apache Group member marcs@znep.com | marc@apache.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 15:03:21 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA10209 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 15:03:21 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from implode.root.com (implode.root.com [198.145.90.17]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA10196 for ; Fri, 24 Apr 1998 15:03:08 -0700 (PDT) (envelope-from root@implode.root.com) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.5/8.8.5) with ESMTP id PAA06860; Fri, 24 Apr 1998 15:01:16 -0700 (PDT) Message-Id: <199804242201.PAA06860@implode.root.com> To: Hugh LaMaster cc: freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-reply-to: Your message of "Fri, 24 Apr 1998 13:21:12 PDT." From: David Greenman Reply-To: dg@root.com Date: Fri, 24 Apr 1998 15:01:16 -0700 Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > >On Fri, 24 Apr 1998, Don Wilde wrote: > >> Yep, that was the first thing I noticed. 5 _Intelligent_ 100Base-T cards >> in theirs. > >I'm not arguing against the Intel cards, but, isn't the performance >on DEC Tulip cards about as good? Any numbers somewhere on this? The DEC card/driver uses about 20% more CPU at the same datarate. -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 16:06:14 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA21363 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 16:06:14 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from vader.cs.berkeley.edu (vader.CS.Berkeley.EDU [128.32.38.234]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA21317; Fri, 24 Apr 1998 16:06:02 -0700 (PDT) (envelope-from asami@vader.cs.berkeley.edu) Received: (from asami@localhost) by vader.cs.berkeley.edu (8.8.7/8.7.3) id QAA17891; Fri, 24 Apr 1998 16:05:58 -0700 (PDT) Date: Fri, 24 Apr 1998 16:05:58 -0700 (PDT) Message-Id: <199804242305.QAA17891@vader.cs.berkeley.edu> To: remy@synx.com CC: dwilde1@ibm.net, don@partsnow.com, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG In-reply-to: <9804241908.aa27782@s3.synx.com> (message from Remy NONNENMACHER on Fri, 24 Apr 1998 19:18:38 +0200 (CEST)) Subject: Re: *** Real Action Item: SPECweb From: asami@FreeBSD.ORG (Satoshi Asami) Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org * From: Remy NONNENMACHER * "Cache volume: NetWare volume spanning six drives, * each with a single 204MB partition" * * This means they used stripping and only the beginning of disks (where * the media rate is the highest). Hard to beat without a well tuned ccd Also don't forget that the seek times go way down if you're only seeking between a few tracks instead of the whole disk. :) Satoshi To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 16:10:24 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA22827 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 16:10:24 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from feral.com (root@[209.54.254.2]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA22531; Fri, 24 Apr 1998 16:09:12 -0700 (PDT) (envelope-from mjacob@feral.com) Received: from feral-gw (mjacob@gw100.feral.com [192.67.166.129]) by feral.com (8.8.6/8.8.6) with SMTP id QAA26472; Fri, 24 Apr 1998 16:08:47 -0700 Message-ID: <35411B7F.445BCF20@feral.com> Date: Fri, 24 Apr 1998 16:08:47 -0700 From: Matthew Jacob Organization: Feral Software X-Mailer: Mozilla 3.04 (X11; I; Linux 2.0.33 i586) MIME-Version: 1.0 To: Satoshi Asami CC: remy@synx.com, dwilde1@ibm.net, don@partsnow.com, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: <199804242305.QAA17891@vader.cs.berkeley.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Satoshi Asami wrote: > > * From: Remy NONNENMACHER > > * "Cache volume: NetWare volume spanning six drives, > * each with a single 204MB partition" > * > * This means they used stripping and only the beginning of disks (where > * the media rate is the highest). Hard to beat without a well tuned ccd > > Also don't forget that the seek times go way down if you're only > seeking between a few tracks instead of the whole disk. :) > > Actually- not necessarily- I've heard that with some of the newer very high density disks that settle time has gone back up to ~2ms or so (while seek time gets < 8ms...). To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 17:02:56 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA00825 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 17:02:56 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from vader.cs.berkeley.edu (vader.CS.Berkeley.EDU [128.32.38.234]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA00740; Fri, 24 Apr 1998 17:02:47 -0700 (PDT) (envelope-from asami@vader.cs.berkeley.edu) Received: (from asami@localhost) by vader.cs.berkeley.edu (8.8.7/8.7.3) id RAA17971; Fri, 24 Apr 1998 17:02:44 -0700 (PDT) Date: Fri, 24 Apr 1998 17:02:44 -0700 (PDT) Message-Id: <199804250002.RAA17971@vader.cs.berkeley.edu> To: mjacob@feral.com CC: remy@synx.com, dwilde1@ibm.net, don@partsnow.com, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG In-reply-to: <35411B7F.445BCF20@feral.com> (message from Matthew Jacob on Fri, 24 Apr 1998 16:08:47 -0700) Subject: Re: *** Real Action Item: SPECweb From: asami@FreeBSD.ORG (Satoshi Asami) Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org * > Also don't forget that the seek times go way down if you're only * > seeking between a few tracks instead of the whole disk. :) * Actually- not necessarily- I've heard that with some of the * newer very high density disks that settle time has gone back * up to ~2ms or so (while seek time gets < 8ms...). That still looks a whole lot faster to me. :) Satoshi To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 17:16:22 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA05665 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 17:16:22 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from misery.sdf.com (misery.sdf.com [204.244.213.32]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id RAA05632; Fri, 24 Apr 1998 17:16:11 -0700 (PDT) (envelope-from tom@sdf.com) Received: from tom by misery.sdf.com with smtp (Exim 1.82 #3) id 0ySsEE-000301-00; Fri, 24 Apr 1998 16:50:02 -0700 Date: Fri, 24 Apr 1998 16:50:01 -0700 (PDT) From: Tom To: Don Wilde cc: Remy NONNENMACHER , don@partsnow.com, freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <3540E0AF.C3B5C576@ibm.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Don Wilde wrote: > Tom wrote: > > > Better off using the hardware RAID0 in a DPT PM334UW. It also has 64MB > > of write-back cache so, it can re-order i/o to be more efficient. > > > > Also, using only tiny bit of each hard drive improves drive level > > caching. Most drives have only a 1MB cache. > > I have a friend at Winchester Systems who's trying to sell me a SCSI-3 > RAID box, and I think they'd be willing to let us borrow one for a test > in return for favorable mention http://www.winsys.com. He says they work > with fast differential 2944's as host cards. Is the DPT faster with the > direct bus connection? No SCSI at all? SCSI-to-SCSI RAID boxes impose latency. Plus the most you can get out a single array is 40MB/s (or 80MB/s with Ultra2). DPT still uses SCSI, but only to talk to the drives. It talks to the host via PCI (132MB/s), not SCSI that the stand-alone RAID boxes use. > Does anybody have experience with these new Compact PCI chassis that > have 8 PCI slots? An obvious bottleneck for anything we can do is going > to be the chassis limitation on most of 4 PCI. I believe the Compaq > Proliant is EISA. Proliant is a series of machines with many different models. Don't buy anything Compaq. Buy a server class motherboard from Intel. I think they make some nice one with lots of PCI slots. Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 17:43:14 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA10653 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 17:43:14 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from cerberus.partsnow.com (gatekeeper.partsnow.com [207.155.26.98] (may be forged)) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA10645 for ; Fri, 24 Apr 1998 17:43:04 -0700 (PDT) (envelope-from don@partsnow.com) Received: (from bin@localhost) by cerberus.partsnow.com (8.8.5/8.6.9) id RAA24427; Fri, 24 Apr 1998 17:42:37 -0700 (PDT) X-Authentication-Warning: cerberus.partsnow.com: bin set sender to using -f Received: from wildeweb(192.168.100.10) by cerberus.partsnow.com via smap (V2.0) id xma024425; Fri, 24 Apr 98 17:42:11 -0700 Message-ID: <3541313C.FBA148E3@partsnow.com> Date: Fri, 24 Apr 1998 17:41:32 -0700 From: Don Wilde Reply-To: don@partsnow.com Organization: Soligen, Incorporated X-Mailer: Mozilla 4.04 [en] (X11; I; FreeBSD 2.2.5-RELEASE i386) MIME-Version: 1.0 To: Tom CC: freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Tom wrote: > SCSI-to-SCSI RAID boxes impose latency. Plus the most you can get out a > single array is 40MB/s (or 80MB/s with Ultra2). > Figured as much, TNX. > DPT still uses SCSI, but only to talk to the drives. It talks to the > host via PCI (132MB/s), not SCSI that the stand-alone RAID boxes use. > Do they use just one chain or do they have the state machine designed to interface each drive separately? > Don't buy anything Compaq. Agreed, I've been burned too. > > Buy a server class motherboard from Intel. I think they make some nice > one with lots of PCI slots. Don't intend to buy, I expect Intel (or somebody) to GIVE me one to play with :) How's this for a scenario: We get an 8-slot passive-PCI chassis (or one of those SuperMicro's that was mentioned before) with a P-II/400 on its master, then put 2 DPT's each with 3-4 small FW-SCSI3 disks and populate the other 5 slots with the Intel intelligent 100/Pro cards. -- oooOOO O O O o * * * * * * o ___ _________ _________ ________ _________ _________ ___==_ V_=_=_DW ===--- Don Wilde [don@PartsNow.com] [http://www.PartsNow.com ] /oo0000oo-oo--oo-ooo---ooo-ooo---ooo-ooo--ooo-ooo---ooo-ooo---ooo-oo--oo  To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 18:12:49 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id SAA14531 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 18:12:49 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA14493; Fri, 24 Apr 1998 18:12:35 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost [127.0.0.1]) by dingo.cdrom.com (8.8.8/8.8.5) with ESMTP id SAA02337; Fri, 24 Apr 1998 18:08:35 -0700 (PDT) Message-Id: <199804250108.SAA02337@dingo.cdrom.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: Bruce Evans cc: mike@smith.net.au, dburr@POBoxes.com, hardware@FreeBSD.ORG, stable@FreeBSD.ORG, Studded@san.rr.com Subject: Re: best wdc0 flags ? In-reply-to: Your message of "Tue, 21 Apr 1998 02:50:39 +1000." <199804201650.CAA05910@godzilla.zeta.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Apr 1998 18:08:35 -0700 From: Mike Smith Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > >> I thought I did, but my oldest accessible drive (all 400MB of it from > >> 4 years ago) supports them. The probe seems to handle any that don't. > > > >OK. Should we make it the default then? > > In -current. Sure. 8) > >> Setting the multi-block flag is not such a good optimization, since it > >> pessimizes throughput on some drives and it increases interrupt latency. > > > >Can you qualify "some drives" again? The overall performance > >improvement in general use is marked, and it decreases interrupt load > >in the DMA case. > > Old drives. Ok. Do we have general consensus then that the defaults should be: - 32-bit transfers. - multi-block 4, if supported by the drive. ??? -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 19:08:32 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA20357 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 19:08:32 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA20340 for ; Fri, 24 Apr 1998 19:08:15 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-240.ca.us.ibm.net [32.100.79.240]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id CAA25246; Sat, 25 Apr 1998 02:07:30 GMT Message-ID: <354141D9.B1B9B36D@ibm.net> Date: Fri, 24 Apr 1998 18:52:25 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: Mike Smith CC: freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: <199804242002.NAA01282@dingo.cdrom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Mike Smith wrote: > These are generally 4 slots and a bridge with 4 more behind. Depending > on the configuration and the hardware in question, more slots is not > necessarily better. I think I agree, although more intelligent controllers on a 132MB bus set is preferable to more disks on an 80MB bus. I think Marc's right in that net throughput appears to be an even bigger issue. Say 2 DPT's, okay. Now, as to the disks themselves. Does anybody know if there's a disk that's intelligent enough to modify it's cacheing algorithms for small-file transfers as opposed to large file work? It seems to me that everybody's jumping on this so-called streaming-video bonanza that's going to be here Real Soon Now, and that's strictly a read-ahead situation. I have had good results with IBM disks, but I've never benched them and I don't know much of the insides. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 19:08:34 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA20373 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 19:08:34 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA20345; Fri, 24 Apr 1998 19:08:19 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-240.ca.us.ibm.net [32.100.79.240]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id CAA25222; Sat, 25 Apr 1998 02:07:25 GMT Message-ID: <35413FC9.55E5AB2C@ibm.net> Date: Fri, 24 Apr 1998 18:43:37 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: Mike Smith CC: Marc Slemko , freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: <199804241831.LAA00871@dingo.cdrom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Mike Smith wrote: > I'm not sure I follow this, but I may be missing a nuance in our > handing of COW. Spinoff from the Bovine Project??? What's COW, never heard of it. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 19:11:56 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA21208 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 19:11:56 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA20954 for ; Fri, 24 Apr 1998 19:11:23 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-240.ca.us.ibm.net [32.100.79.240]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id CAA98880; Sat, 25 Apr 1998 02:11:12 GMT Message-ID: <3541464B.467965E2@ibm.net> Date: Fri, 24 Apr 1998 19:11:23 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: Marc@ibm.net, Slemko@ibm.net CC: freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Marc Slemko wrote: > > Because of the design of Apache (more so on servers with more config > info, thousands of vhosts, etc.) where the child processes are forked > from the parent, you can end up with a lot of COW pages. > Marc - Could you give me a little insight into why you think Zeus is better? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 19:14:34 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA21629 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 19:14:34 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA21607 for ; Fri, 24 Apr 1998 19:14:18 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-240.ca.us.ibm.net [32.100.79.240]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id CAA34754; Sat, 25 Apr 1998 02:14:14 GMT Message-ID: <35414701.F1065C80@ibm.net> Date: Fri, 24 Apr 1998 19:14:25 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: Marc Slemko CC: freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Could you give us some insight into what makes Zeus better, Marc? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 19:22:43 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA23015 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 19:22:43 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from pendor.McKusick.COM (root@c987316-a.frmt1.sfba.home.com [24.1.79.174]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA22989; Fri, 24 Apr 1998 19:22:24 -0700 (PDT) (envelope-from benco@pendor.McKusick.COM) Received: from localhost (benco@localhost [127.0.0.1]) by pendor.McKusick.COM (8.8.8/8.8.7) with SMTP id TAA25823; Fri, 24 Apr 1998 19:22:19 -0700 (PDT) Message-Id: <199804250222.TAA25823@pendor.McKusick.COM> X-Authentication-Warning: pendor.McKusick.COM: benco@localhost [127.0.0.1] didn't use HELO protocol To: freebsd-advocacy@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-reply-to: Your message of "Fri, 24 Apr 1998 18:43:37 PDT." <35413FC9.55E5AB2C@ibm.net> Date: Fri, 24 Apr 1998 19:22:18 -0700 From: Ben Cottrell Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998 18:43:37 -0700, Don Wilde wrote: > Spinoff from the Bovine Project??? What's COW, never heard of it. Haven't been following this discussion, but I did want to point out that COW is Copy On Write. ~Ben To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 20:31:21 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id UAA28999 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 20:31:21 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA28842 for ; Fri, 24 Apr 1998 20:30:55 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-178.ca.us.ibm.net [32.100.79.178]) by out1.ibm.net (8.8.5/8.6.9) with ESMTP id DAA76380; Sat, 25 Apr 1998 03:30:45 GMT Message-ID: <354158BF.A4F1E284@ibm.net> Date: Fri, 24 Apr 1998 20:30:07 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: Marc Slemko CC: freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org First, I've been asked to choose one -list or another, so let's all limit this to -hardware and not -advocacy, since it's really getting too deep for our enthusiastic types over there. I think our answer to the 'stable storage' rule you quoted is that we get a couple of 512M FlashROM disks for the data files. That satisfies both the spirit and the letter of the rules. This is right in line with the 'Philosophy' section. I would definitely recommend Solid State storage over rotating media for a webserver, especially directly wired to the PCI memory map if we can get it. Notice they didn't say cost had to be a consideration! ;) I see that a 512M flash Drive is $12849 each. A battery backed DRAM drive of 256M is $1695+the DRAM cost. It has 12 hours of backup; I'd say that is 'stable' enough for a webserver which will be on 24x7 duty. Max read-rate speed is only 10MB, sustained transfer is less, but still pretty fast and the access time is quoted as "<0.1ms" for either. I'm certain there would be no comparison between these and any disk drive!!! We use disks only for logs, volatile data files and non-essential system files. As I see it, we are allowed to use RAM to hold as many server daemons as we want, and it doesn't say anything about a requirement for CGI. Am I correct in assuming that this is purely HTML and graphics files? Back to disk controllers for a second. I see in my ICS catalog that there is a DPT controller (PM3334/UW3) that supports 3 SCSI bus strings in RAID 0. Comments? On the earlier question of Apache vs. Zeus, I'm still inclined to stick with Apache. Again my reasoning is that we are out to promote freeware, and Apache is a known name even to the Wall Street Journal. I'm going to go back and read some of the earlier SPEC results and see what else is out there for other single-processor machines, but I'll bet using ROM/B-DRAM disks will multiply our throughput up to the point where we're back to net performance as _the_ issue. Speaking of which, the price differential between the normal Intel 10/100 and the SERVER version is $484.00. I'd say there's a 960 in there... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 22:20:25 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA08974 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 22:20:25 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from portal.net.au (galley.portal.net.au [202.12.71.7]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id WAA08951 for ; Fri, 24 Apr 1998 22:20:14 -0700 (PDT) (envelope-from matt@portal.net.au) Received: (from matt@localhost) by portal.net.au (8.8.8/8.8.8) id OAA01718 for freebsd-hardware@freebsd.org; Sat, 25 Apr 1998 14:50:00 +0930 (CST) From: Matt Baker Message-Id: <199804250520.OAA01718@portal.net.au> Subject: Re: *** Real Action Item: SPECweb To: freebsd-hardware@FreeBSD.ORG Date: Sat, 25 Apr 1998 14:49:59 +0930 (CST) In-Reply-To: <354158BF.A4F1E284@ibm.net> from "Don Wilde" at Apr 24, 98 08:30:07 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > On the earlier question of Apache vs. Zeus, I'm still inclined to stick > with Apache. Again my reasoning is that we are out to promote freeware, > and Apache is a known name even to the Wall Street Journal. I'm going to > go back and read some of the earlier SPEC results and see what else is > out there for other single-processor machines, but I'll bet using > ROM/B-DRAM disks will multiply our throughput up to the point where > we're back to net performance as _the_ issue. Has anyone done a comparison test between using Apache by itself, and a Apache server with a Squid frontend? Squid can run in accelarator mode for web servers, and certainly seems quicker than Apache at serving pages. Matt. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Fri Apr 24 23:28:43 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA15124 for freebsd-hardware-outgoing; Fri, 24 Apr 1998 23:28:43 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA15119 for ; Fri, 24 Apr 1998 23:28:40 -0700 (PDT) (envelope-from gibbs@narnia.plutotech.com) Received: (from gibbs@localhost) by narnia.plutotech.com (8.8.8/8.7.3) id AAA04043; Sat, 25 Apr 1998 00:24:46 -0600 (MDT) Date: Sat, 25 Apr 1998 00:24:46 -0600 (MDT) From: "Justin T. Gibbs" Message-Id: <199804250624.AAA04043@narnia.plutotech.com> To: dwilde1@ibm.net cc: hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb Newsgroups: pluto.freebsd.hardware In-Reply-To: <199804242002.NAA01282@dingo.cdrom.com> <354141D9.B1B9B36D@ibm.net> User-Agent: tin/pre-1.4-971204 (UNIX) (FreeBSD/3.0-CURRENT (i386)) Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Say 2 DPT's, okay. Now, as to the disks themselves. Does anybody know if > there's a disk that's intelligent enough to modify it's cacheing > algorithms for small-file transfers as opposed to large file work? It > seems to me that everybody's jumping on this so-called streaming-video > bonanza that's going to be here Real Soon Now, and that's strictly a > read-ahead situation. It is only a read-ahead/write behind issue if you are dealing with a single stream of video that was recorded in a relatively contiguous manner. When you have to record or play back multiple 25mb (DVCPro), 50mb (DVCPro-50), 270mb (Uncompressed 525 SDTV with vertical interval), or even 360mb (16x9 525 Uncompressed or 720P or 1080i HDTV compressed) you are really seek bound. Luckily most multi-stream applications are in the broacast realm where compressed formats are acceptable (DVCPro 25 or 50), but then they want even more streams, meaning more fragmenttation, and more seeks. Of couse it is possible to do all of this, using FreeBSD even: http://www.plutotech.com/ We now return you to your regularly scheduled hardware discussion. 8-) -- Justin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sat Apr 25 01:00:19 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA22685 for freebsd-hardware-outgoing; Sat, 25 Apr 1998 01:00:19 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from misery.sdf.com (misery.sdf.com [204.244.213.32]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id BAA22676 for ; Sat, 25 Apr 1998 01:00:13 -0700 (PDT) (envelope-from tom@sdf.com) Received: from tom by misery.sdf.com with smtp (Exim 1.82 #3) id 0ySzTR-0003FN-00; Sat, 25 Apr 1998 00:34:13 -0700 Date: Sat, 25 Apr 1998 00:34:12 -0700 (PDT) From: Tom To: Don Wilde cc: freebsd-hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb In-Reply-To: <3541313C.FBA148E3@partsnow.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 24 Apr 1998, Don Wilde wrote: > Do they use just one chain or do they have the state machine designed to > interface each drive separately? How does anyone know how the implementation of any piece of hardware works? You don't. You instead try it and measure it. Simon has some posted some results on i/o operations per second that he has measured (4000/s I believe). .. > How's this for a scenario: > We get an 8-slot passive-PCI chassis (or one of those SuperMicro's that was > mentioned before) with a P-II/400 on its master, then put 2 DPT's each with 3-4 > small FW-SCSI3 disks and populate the other 5 slots with the Intel intelligent > 100/Pro cards. One disk per SCSI channel? I think it is more important to look at the array config (proabably RAID 1), and strip size settings, plus the various tuning paramters. Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message From owner-freebsd-hardware Sat Apr 25 07:42:03 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA29110 for freebsd-hardware-outgoing; Sat, 25 Apr 1998 07:42:03 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from out2.ibm.net (out2.ibm.net [165.87.194.229]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA29082 for ; Sat, 25 Apr 1998 07:41:58 -0700 (PDT) (envelope-from dwilde1@ibm.net) Received: from ibm.net (slip-32-100-79-90.ca.us.ibm.net [32.100.79.90]) by out2.ibm.net (8.8.5/8.6.9) with ESMTP id OAA149846; Sat, 25 Apr 1998 14:41:53 GMT Message-ID: <3541DDC8.6AB95C12@ibm.net> Date: Sat, 25 Apr 1998 05:57:44 -0700 From: Don Wilde Reply-To: dwilde1@ibm.net X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: "Justin T. Gibbs" CC: hardware@FreeBSD.ORG Subject: Re: *** Real Action Item: SPECweb References: <199804250624.AAA04043@narnia.plutotech.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org