From owner-freebsd-hardware Mon Feb 17 11:38:00 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA09372 for hardware-outgoing; Mon, 17 Feb 1997 11:38:00 -0800 (PST) Received: from gvr.win.tue.nl (root@gvr.win.tue.nl [131.155.210.19]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA09338 for ; Mon, 17 Feb 1997 11:37:46 -0800 (PST) Received: (from guido@localhost) by gvr.win.tue.nl (8.8.5/8.8.2) id UAA24903; Mon, 17 Feb 1997 20:37:11 +0100 (MET) From: Guido van Rooij Message-Id: <199702171937.UAA24903@gvr.win.tue.nl> Subject: Re: Problems with 3COM509 Etherlink III on FreeBSD 2.1.6 In-Reply-To: <3.0.32.19970211145720.006a0544@mail.microweb.com> from "David A. Zimmerman" at "Feb 11, 97 02:57:25 pm" To: daz@sssystems.com (David A. Zimmerman) Date: Mon, 17 Feb 1997 20:37:11 +0100 (MET) Cc: freebsd-hardware@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk David A. Zimmerman wrote: > I am not a UNIX expert by an strech but have had a good deal of experience > and feel that I have tried just about everything I can think of over the > last two weeks. Would anyone be kind enough to offer any insight into a > few other things I might try to get this machine on the Net? What am I > missing? How can I diagnose this problem? > What are you using? The AUI, UTP or BNC connector? Look at the manpage to check which links to specify for ifconfig...: -link0 Use the BNC port (default). link0 -link1 Use the AUI port. link0 link1 Use the UTP port. -Guido From owner-freebsd-hardware Tue Feb 18 06:59:08 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA08078 for hardware-outgoing; Tue, 18 Feb 1997 06:59:08 -0800 (PST) Received: from bunyip.cc.uq.edu.au (daemon@bunyip.cc.uq.edu.au [130.102.2.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA08065 for ; Tue, 18 Feb 1997 06:59:03 -0800 (PST) Received: (from daemon@localhost) by bunyip.cc.uq.edu.au (8.8.5/8.8.5) id AAA19174; Wed, 19 Feb 1997 00:58:48 +1000 Received: by ogre.devetir.qld.gov.au (8.7.5/DEVETIR-E0.3a) id WAA12652; Tue, 18 Feb 1997 22:01:24 +1000 (EST) Date: Tue, 18 Feb 1997 22:01:24 +1000 (EST) From: Stephen McKay Message-Id: <199702181201.WAA12652@ogre.devetir.qld.gov.au> To: asami@vader.cs.berkeley.edu (Satoshi Asami) cc: freebsd-hardware@freebsd.org, syssgm@devetir.qld.gov.au Subject: Re: 64 MB ECC or 128 MB non ECC ? X-Newsreader: NN version 6.5.0 #1 (NOV) Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk asami@vader.cs.berkeley.edu (Satoshi Asami) wrote: > * Well, I'm glad I've got some cache to enable since without cache the kernel > * compile takes 15 times as long. Yes, my ECC, nocache test took 97 minutes! > * Yikes! The final result is that ECC is 12% slower than PARITY with no cache. > * (user+sys time only). This is in agreement with those people who predicted > * 10-15% main memory slow down, but, as noted above, reduces to 1% with both > >Thanks for verifying it. I've found out something new that means that what I've "verified" might not be what you thought I was verifying. Intel's TXC documentation states that you have to reduce your read accesses to x333 if using ECC. (There is a one clock lead-off penalty too.) Since I'm already running my FPM at x333 there would be (next to) no penalty there. Only EDO users would suffer. Looks like the read-modify-write penalty (as mitigated by the TXC write reassembly buffers) was what I was really measuring. > * caches enabled. 1% is no pain at all. > >Well that depends, if you are X-user-switch-around-between-20-windows >type of guy, you can actually feel the 10% slowdown as you use the >machine. I switched back to parity because I couldn't stand it. ;) Hmm. I beat my machines until they beg for mercy. :-) I haven't run X on the new box yet as I'm still experimenting with weird stuff (like old SCSI disks that lock up the SCSI bus), and I can't afford a second good monitor. If I never run it for real without ECC, I'll never know what I'm missing. :-) Stephen. From owner-freebsd-hardware Tue Feb 18 09:20:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA16369 for hardware-outgoing; Tue, 18 Feb 1997 09:20:11 -0800 (PST) Received: from acc0.elvisti.kiev.ua (acc0.elvisti.kiev.ua [193.125.28.132]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA16102; Tue, 18 Feb 1997 09:15:50 -0800 (PST) Received: from office.elvisti.kiev.ua (office.elvisti.kiev.ua [193.125.28.129]) by acc0.elvisti.kiev.ua (8.8.5/8.8.ElVisti) with ESMTP id TAA16339; Tue, 18 Feb 1997 19:14:45 +0200 (EET) Received: from gw.elvisti.kiev.ua (gw.elvisti.kiev.ua [193.125.28.140]) by office.elvisti.kiev.ua (8.6.12/8.ElVisti) with ESMTP id TAA24012; Tue, 18 Feb 1997 19:14:45 +0200 Received: from pinta.elvisti.kiev.ua (pinta.elvisti.kiev.ua [193.125.28.145]) by gw.elvisti.kiev.ua (8.8.5/8.6.12) with ESMTP id TAA23991; Tue, 18 Feb 1997 19:14:44 +0200 (EET) Received: from pinta.elvisti.kiev.ua (pinta.elvisti.kiev.ua [193.125.28.145]) by pinta.elvisti.kiev.ua (8.8.5/8.7.3) with SMTP id TAA19833; Tue, 18 Feb 1997 19:14:44 +0200 (EET) Message-ID: <3309E383.2D857063@elvisti.kiev.ua> Date: Tue, 18 Feb 1997 17:14:43 +0000 From: Andrei Biryukov Organization: ElVisti X-Mailer: Mozilla 3.01Gold (X11; I; FreeBSD 2.2-ALPHA i386) MIME-Version: 1.0 To: FreeBSD-hackers@freebsd.org, FreeBSD-current@freebsd.org, FreeBSD-hardware@freebsd.org, FreeBSD-quections@freebsd.org Subject: is MOXA C-204 supported ? References: <199702181627.JAA11155@phaeton.artisoft.com> Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 8bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I've got a question. Does FreeBSD support the Multiport serial sync Óard MOXA C-204 or not ? -- Biryukov Andrei IC ElVisti E-mail amb@elvisti.kiev.ua From owner-freebsd-hardware Tue Feb 18 13:27:35 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA02550 for hardware-outgoing; Tue, 18 Feb 1997 13:27:35 -0800 (PST) Received: from fallout.campusview.indiana.edu (fallout.campusview.indiana.edu [149.159.1.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA02542 for ; Tue, 18 Feb 1997 13:27:30 -0800 (PST) Received: from localhost (jfieber@localhost) by fallout.campusview.indiana.edu (8.8.5/8.8.5) with SMTP id QAA00570 for ; Tue, 18 Feb 1997 16:27:22 -0500 (EST) Date: Tue, 18 Feb 1997 16:27:22 -0500 (EST) From: John Fieber To: hardware@freebsd.org Subject: Motherboard doesn't like tape drive Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk [I asked this a month or two ago and no explaination or solution came out of it, so I'll try again. Hopefully someone will have a flash of insight...] Before: 80486dx33 motherboard + adaptec 1542C + archive viper 150 tape drive. When writing to the tape, with , the tape streams nicely for about 100KBytes/sec throughput. After: ASUS P/I-P55T2P4 Pentum 100 motherboard + ASUS SC200 + archive viper 150 tape drive. I've tried every trick in the book, every snazzy buffering tape streaming enhancement gizmo and the thing just won't stream anymore! I get about 60KBytes/sec throughput on a good day. I put the old adaptec 1542C controller back in but that made no difference. This is really aggrivating! It takes *forever* to backup a 1 gig disk at 60KBytes/sec, never mind the wear and tear on the drive of starting and stopping so often. I've been tracking the RELENG_2_2 kernel sources and observed no change in the situation in the past month or two. The video card is the only other component of the system that changed in the upgrade. The hard drive, sound card, network card, amount of RAM are all the same as before. -john From owner-freebsd-hardware Tue Feb 18 14:21:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA05555 for hardware-outgoing; Tue, 18 Feb 1997 14:21:53 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA05541 for ; Tue, 18 Feb 1997 14:21:38 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.2/8.7.3) id IAA06984; Wed, 19 Feb 1997 08:51:26 +1030 (CST) From: Michael Smith Message-Id: <199702182221.IAA06984@genesis.atrad.adelaide.edu.au> Subject: Re: Motherboard doesn't like tape drive In-Reply-To: from John Fieber at "Feb 18, 97 04:27:22 pm" To: jfieber@indiana.edu (John Fieber) Date: Wed, 19 Feb 1997 08:51:25 +1030 (CST) Cc: hardware@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk John Fieber stands accused of saying: > > After: ASUS P/I-P55T2P4 Pentum 100 motherboard + ASUS SC200 > + archive viper 150 tape drive. I've tried every trick > in the book, every snazzy buffering tape streaming > enhancement gizmo and the thing just won't stream anymore! > I get about 60KBytes/sec throughput on a good day. I put the > old adaptec 1542C controller back in but that made no difference. IDE or SCSI disk? The Archive has a rather small buffer (24K IIRC), and it's quite possible that it'll starve in the time that it takes your system to talk to an IDE disk. If you have any of the go-faster flags turned on on the IDE disk, try turning them off. (Particularly the multi-block stuff). > -john -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hardware Tue Feb 18 14:26:51 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA05842 for hardware-outgoing; Tue, 18 Feb 1997 14:26:51 -0800 (PST) Received: from fallout.campusview.indiana.edu (fallout.campusview.indiana.edu [149.159.1.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA05834 for ; Tue, 18 Feb 1997 14:26:42 -0800 (PST) Received: from localhost (jfieber@localhost) by fallout.campusview.indiana.edu (8.8.5/8.8.5) with SMTP id RAA00699; Tue, 18 Feb 1997 17:26:30 -0500 (EST) Date: Tue, 18 Feb 1997 17:26:30 -0500 (EST) From: John Fieber To: Michael Smith cc: hardware@freebsd.org Subject: Re: Motherboard doesn't like tape drive In-Reply-To: <199702182221.IAA06984@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 19 Feb 1997, Michael Smith wrote: > IDE or SCSI disk? IDE? What's that? :) No, I don't have any IDE devices. The IDE controller is disabled in the motherboard BIOS. Actually, now that I think about it, I did have an IDE disk attached to the old motherboard, but the tape drive worked fine in that setup. -john From owner-freebsd-hardware Tue Feb 18 14:47:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA06778 for hardware-outgoing; Tue, 18 Feb 1997 14:47:07 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA06764 for ; Tue, 18 Feb 1997 14:46:56 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.2/8.7.3) id JAA07149; Wed, 19 Feb 1997 09:16:44 +1030 (CST) From: Michael Smith Message-Id: <199702182246.JAA07149@genesis.atrad.adelaide.edu.au> Subject: Re: Motherboard doesn't like tape drive In-Reply-To: from John Fieber at "Feb 18, 97 05:26:30 pm" To: jfieber@indiana.edu (John Fieber) Date: Wed, 19 Feb 1997 09:16:43 +1030 (CST) Cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk John Fieber stands accused of saying: > On Wed, 19 Feb 1997, Michael Smith wrote: > > > IDE or SCSI disk? > > IDE? What's that? :) OK, that's what I like to hear 8) Do you have "FAILSAFE" in your kernel config? Check with Stefan to make sure that there's nothing funny happening with your drive wrt. disconnect etc. I haven't had any trouble with my Viper 150 under 2.2 on a Bustek BT542. -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hardware Tue Feb 18 14:57:24 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA07414 for hardware-outgoing; Tue, 18 Feb 1997 14:57:24 -0800 (PST) Received: from fallout.campusview.indiana.edu (fallout.campusview.indiana.edu [149.159.1.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA07397 for ; Tue, 18 Feb 1997 14:56:58 -0800 (PST) Received: from localhost (jfieber@localhost) by fallout.campusview.indiana.edu (8.8.5/8.8.5) with SMTP id RAA00747; Tue, 18 Feb 1997 17:56:40 -0500 (EST) Date: Tue, 18 Feb 1997 17:56:39 -0500 (EST) From: John Fieber To: Michael Smith cc: hardware@freebsd.org Subject: Re: Motherboard doesn't like tape drive In-Reply-To: <199702182246.JAA07149@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 19 Feb 1997, Michael Smith wrote: > Do you have "FAILSAFE" in your kernel config? Nope. Should I? The one tunable parameter on the archive viper is the buffer disconnect size. It is currently at 16K (factory default), but can be adjusted between 2K and 32K. Could that be a factor here? -john From owner-freebsd-hardware Tue Feb 18 15:11:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA08181 for hardware-outgoing; Tue, 18 Feb 1997 15:11:13 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA08173 for ; Tue, 18 Feb 1997 15:10:59 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.2/8.7.3) id JAA07385; Wed, 19 Feb 1997 09:40:43 +1030 (CST) From: Michael Smith Message-Id: <199702182310.JAA07385@genesis.atrad.adelaide.edu.au> Subject: Re: Motherboard doesn't like tape drive In-Reply-To: from John Fieber at "Feb 18, 97 05:56:39 pm" To: jfieber@indiana.edu (John Fieber) Date: Wed, 19 Feb 1997 09:40:42 +1030 (CST) Cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk John Fieber stands accused of saying: > On Wed, 19 Feb 1997, Michael Smith wrote: > > > Do you have "FAILSAFE" in your kernel config? > > Nope. Should I? No; the ncr driver behaves more conservatively when it's defined, I thought that might be a problem. > The one tunable parameter on the archive viper is the buffer > disconnect size. It is currently at 16K (factory default), but > can be adjusted between 2K and 32K. Could that be a factor here? Indeed. Is that the high-water or low-water mark for disconnect? Try fiddling it and see 8) > -john -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hardware Tue Feb 18 15:22:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA08694 for hardware-outgoing; Tue, 18 Feb 1997 15:22:56 -0800 (PST) Received: from fallout.campusview.indiana.edu (fallout.campusview.indiana.edu [149.159.1.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA08668 for ; Tue, 18 Feb 1997 15:21:51 -0800 (PST) Received: from localhost (jfieber@localhost) by fallout.campusview.indiana.edu (8.8.5/8.8.5) with SMTP id SAA00810; Tue, 18 Feb 1997 18:21:12 -0500 (EST) Date: Tue, 18 Feb 1997 18:21:12 -0500 (EST) From: John Fieber To: Michael Smith cc: hardware@freebsd.org Subject: Re: Motherboard doesn't like tape drive In-Reply-To: <199702182310.JAA07385@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 19 Feb 1997, Michael Smith wrote: > > The one tunable parameter on the archive viper is the buffer > > disconnect size. It is currently at 16K (factory default), but > > can be adjusted between 2K and 32K. Could that be a factor here? > > Indeed. Is that the high-water or low-water mark for disconnect? >From the manual: "The buffer disconnect size sets the maximum number of bytes that can be sent over the SCSI bus during a single data transfer phase. During lengthy data transfers, this feature periodically frees the bus for other operations; thus, the disocnnect size can affect overall system performance where more than two SCSI devices must arbitrate for bus time. Optimum disconnect size is a function of both the number of devices and the speed at which they independently process data burts." On the jumper chart, there is a footnote saying that "Disconnect size setting 16K is minimum for using the Copy command." > Try fiddling it and see 8) My computer hold up one end of some boards I call a bookshelf. Hardware fiddling is always a hassle. :( -john From owner-freebsd-hardware Tue Feb 18 15:50:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA10761 for hardware-outgoing; Tue, 18 Feb 1997 15:50:50 -0800 (PST) Received: from earth.infinetconsulting.com (earth.infinetconsulting.com [207.23.43.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA10750 for ; Tue, 18 Feb 1997 15:50:32 -0800 (PST) Received: (from lenc@localhost) by earth.infinetconsulting.com (8.8.3/8.6.12) id QAA28853; Tue, 18 Feb 1997 16:04:15 -0800 (PST) Date: Tue, 18 Feb 1997 16:04:14 -0800 (PST) From: Leonard Chua To: PB , "Luc.LEWY" , freebsd-hardware@freebsd.org Subject: Re: Adaptec AHA1520A help needed. found it! In-Reply-To: <199702131308.NAA09130@wave.campus.luth.se> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Adaptec has an interactive FaxFor service. Call (408) 957-7150 and order document# 34025. I believe the FaxFor service is for anywhere in the world. If not, email sales@corp.adaptec.com for a hardcopy. Cheers! Len. On Thu, 13 Feb 1997, PB wrote: > Leonard Chua wrote: > >Hi, > > I've this second hand AHA1520A card with absolutely no manuals > >and no software. The software I can get from adaptec's web site. > >But I need to know the jumper settings though. Does anyone have > >a manual lying around that I can get? > >Much thanks in advance. > > > >Cheers. > >Len. > > > > > > Hi,! > > If you get any info, on this please send me a copy.. > I'll check with my computer society, seems like they got one in our CD-R > machine. > > /Peter > > Email: peterb@ludd.luth.se > > From owner-freebsd-hardware Tue Feb 18 16:45:40 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA15044 for hardware-outgoing; Tue, 18 Feb 1997 16:45:40 -0800 (PST) Received: (from jmb@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA15022; Tue, 18 Feb 1997 16:45:14 -0800 (PST) From: "Jonathan M. Bresler" Message-Id: <199702190045.QAA15022@freefall.freebsd.org> Subject: Re: Motherboard doesn't like tape drive To: jfieber@indiana.edu (John Fieber) Date: Tue, 18 Feb 1997 16:45:13 -0800 (PST) Cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org In-Reply-To: from "John Fieber" at Feb 18, 97 06:21:12 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk John Fieber wrote: > > On Wed, 19 Feb 1997, Michael Smith wrote: > > > > The one tunable parameter on the archive viper is the buffer > > > disconnect size. It is currently at 16K (factory default), but > > > can be adjusted between 2K and 32K. Could that be a factor here? > > > > Indeed. Is that the high-water or low-water mark for disconnect? > > >From the manual: > > "The buffer disconnect size sets the maximum number of bytes that > can be sent over the SCSI bus during a single data transfer > phase. During lengthy data transfers, this feature periodically > frees the bus for other operations; thus, the disocnnect size can > affect overall system performance where more than two SCSI > devices must arbitrate for bus time. Optimum disconnect size is > a function of both the number of devices and the speed at which > they independently process data burts." the problem may be that the tape is straved for data, so it stops writing and has to seek before starting to write again. could hte disks be holding the bus for too long? can you change the "buffer disconnect size" on the disks? jmb From owner-freebsd-hardware Tue Feb 18 16:52:48 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA15738 for hardware-outgoing; Tue, 18 Feb 1997 16:52:48 -0800 (PST) Received: from starbase.globalpc.net ([207.211.100.102]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA15732 for ; Tue, 18 Feb 1997 16:52:45 -0800 (PST) Received: (from agonzalez@localhost) by starbase.globalpc.net (8.8.3/8.8.3) id SAA04590; Tue, 18 Feb 1997 18:56:39 -0600 (CST) Date: Tue, 18 Feb 1997 18:56:39 -0600 (CST) From: Adrian Gonzalez To: freebsd-hardware@freebsd.org Subject: Digiboard PC/16e problem Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi I've been trying to get a Digiboard PC/16e multiport card to work with FreeBSD 2.1.6 and I get the following error message at bootup: real memory = 16777216 (16384K bytes) avail memory = 14946304 (14596K bytes) dgb0: PC/Xe 64K dgb0 at 0x220-0x223 maddr 0xf0000 msize 65536 on isa dgb0: 4th memory test (BIOS load) fails dgb0: BIOS download failed The dip switches on the board are set for no irq, port 0x220, ram at 0xf0000. I've tried moving the board to 0xe0000, and that won't do it for me either. If you have any idea what could be wrong, please let me know. I'd really like to get this thing working because I have this spare digiboard lying around unused. thanks in advance -Adrian Gonzalez From owner-freebsd-hardware Tue Feb 18 17:16:21 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA18278 for hardware-outgoing; Tue, 18 Feb 1997 17:16:21 -0800 (PST) Received: from fallout.campusview.indiana.edu (fallout.campusview.indiana.edu [149.159.1.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA18257 for ; Tue, 18 Feb 1997 17:16:11 -0800 (PST) Received: from localhost (jfieber@localhost) by fallout.campusview.indiana.edu (8.8.5/8.8.5) with SMTP id UAA01088; Tue, 18 Feb 1997 20:15:34 -0500 (EST) Date: Tue, 18 Feb 1997 20:15:34 -0500 (EST) From: John Fieber To: "Jonathan M. Bresler" cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org Subject: Re: Motherboard doesn't like tape drive In-Reply-To: <199702190045.QAA15022@freefall.freebsd.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 18 Feb 1997, Jonathan M. Bresler wrote: > the problem may be that the tape is straved for data, so it stops > writing and has to seek before starting to write again. Then I would think that dd'ing /dev/zero to the tape would keep it streaming, but it won't. :( > could hte disks be holding the bus for too long? > can you change the "buffer disconnect size" on the disks? Don't know. There is one hard drive, a Fujitsu M1606S (bought from Rod Grimes, so it should be "FreeBSD Certified"), and the only jumpers are for SCSI ID and spindle motor. Is this something that could be set through software? There is also a Toshiba 3501 CD-ROM drive on the bus. I should note, that at one point I had the hard drive alone on the SC200 controller, and the Viper alone on the Adaptec 1542C, and the result was the same: no streaming. I've tried different cables as well. It has also crossed my mind that the tape drive may just be broken, and it is coincidence that it happened right after (or during) a motherboard swap. Anybody know typical failure symptoms for the Archive Viper? I don't get read or write errors, just no streaming. -john From owner-freebsd-hardware Tue Feb 18 18:09:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA21753 for hardware-outgoing; Tue, 18 Feb 1997 18:09:28 -0800 (PST) Received: from lserver.infoworld.com (lserver.infoworld.com [192.216.48.4]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA21746 for ; Tue, 18 Feb 1997 18:09:20 -0800 (PST) Received: from IWND1.infoworld.com (iwnd1.infoworld.com [192.216.49.131]) by lserver.infoworld.com (8.8.5/8.8.5/GNAC-GW-2.1) with SMTP id SAA18329; Tue, 18 Feb 1997 18:08:23 -0800 (PST) Received: by IWND1.infoworld.com(Lotus SMTP MTA v1.05 (305.3 1-15-1997)) id 88256443.000B67EE ; Tue, 18 Feb 1997 18:04:35 -0700 X-Lotus-FromDomain: IW @ IWP From: "Brett Glass at POST-IW1" To: agonzalez@starbase.globalpc.net, freebsd-hardware@freebsd.org Message-ID: <88256443.000B4ADC.00@IWND1.infoworld.com> Date: Tue, 18 Feb 1997 18:57:00 -0700 Subject: Digiboard PC/16e problem Mime-Version: 1.0 Content-type: text/plain; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk For some reason, the FreeBSD drivers don't work with many of the Digi products. I have an X/em that locks up at the same point. From owner-freebsd-hardware Tue Feb 18 19:42:00 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA28055 for hardware-outgoing; Tue, 18 Feb 1997 19:42:00 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA28048 for ; Tue, 18 Feb 1997 19:41:55 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.2/8.7.3) id OAA09307; Wed, 19 Feb 1997 14:11:08 +1030 (CST) From: Michael Smith Message-Id: <199702190341.OAA09307@genesis.atrad.adelaide.edu.au> Subject: Re: Motherboard doesn't like tape drive In-Reply-To: from John Fieber at "Feb 18, 97 08:15:34 pm" To: jfieber@indiana.edu (John Fieber) Date: Wed, 19 Feb 1997 14:11:07 +1030 (CST) Cc: jmb@freefall.freebsd.org, msmith@atrad.adelaide.edu.au, hardware@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk John Fieber stands accused of saying: > On Tue, 18 Feb 1997, Jonathan M. Bresler wrote: > > > the problem may be that the tape is straved for data, so it stops > > writing and has to seek before starting to write again. > > Then I would think that dd'ing /dev/zero to the tape would keep > it streaming, but it won't. :( That might happen if it gets its data out to the tape before your system has got back to it with more. 8( > It has also crossed my mind that the tape drive may just be > broken, and it is coincidence that it happened right after (or > during) a motherboard swap. Anybody know typical failure > symptoms for the Archive Viper? I don't get read or write > errors, just no streaming. Don't think so; the only symptoms I've seen have been "dead, does nothing", and "lots of errors". > -john -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hardware Tue Feb 18 19:46:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA28252 for hardware-outgoing; Tue, 18 Feb 1997 19:46:13 -0800 (PST) Received: from fallout.campusview.indiana.edu (fallout.campusview.indiana.edu [149.159.1.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA28236 for ; Tue, 18 Feb 1997 19:45:52 -0800 (PST) Received: from localhost (jfieber@localhost) by fallout.campusview.indiana.edu (8.8.5/8.8.5) with SMTP id WAA01939; Tue, 18 Feb 1997 22:45:25 -0500 (EST) Date: Tue, 18 Feb 1997 22:45:24 -0500 (EST) From: John Fieber To: Michael Smith cc: jmb@freefall.freebsd.org, hardware@freebsd.org Subject: Re: Motherboard doesn't like tape drive In-Reply-To: <199702190341.OAA09307@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 19 Feb 1997, Michael Smith wrote: > John Fieber stands accused of saying: > > On Tue, 18 Feb 1997, Jonathan M. Bresler wrote: > > > > > the problem may be that the tape is straved for data, so it stops > > > writing and has to seek before starting to write again. > > > > Then I would think that dd'ing /dev/zero to the tape would keep > > it streaming, but it won't. :( > > That might happen if it gets its data out to the tape before your > system has got back to it with more. 8( Why might this happen on Pentium 100 with a PCI SCSI controller where my crusty old 486 with an ISA SCSI controller could keep the drive well fed? Inquiring minds want to know. :) Well, I'll probably worry about this problem (again) in another month or so when its time for another full backup. -john From owner-freebsd-hardware Tue Feb 18 19:52:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA28876 for hardware-outgoing; Tue, 18 Feb 1997 19:52:53 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA28866 for ; Tue, 18 Feb 1997 19:52:43 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.2/8.7.3) id OAA09408; Wed, 19 Feb 1997 14:22:13 +1030 (CST) From: Michael Smith Message-Id: <199702190352.OAA09408@genesis.atrad.adelaide.edu.au> Subject: Re: Motherboard doesn't like tape drive In-Reply-To: from John Fieber at "Feb 18, 97 10:45:24 pm" To: jfieber@indiana.edu (John Fieber) Date: Wed, 19 Feb 1997 14:22:12 +1030 (CST) Cc: msmith@atrad.adelaide.edu.au, jmb@freefall.freebsd.org, hardware@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk John Fieber stands accused of saying: > > Why might this happen on Pentium 100 with a PCI SCSI controller > where my crusty old 486 with an ISA SCSI controller could keep > the drive well fed? No idea. Maybe you work your P100 harder than you used to work your 486. 8) Honestly, I don't know. I would hope that > -john -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hardware Wed Feb 19 12:00:03 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA15066 for hardware-outgoing; Wed, 19 Feb 1997 12:00:03 -0800 (PST) Received: from Sisyphos.MI.Uni-Koeln.DE (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA15017; Wed, 19 Feb 1997 11:59:26 -0800 (PST) Received: from x14.mi.uni-koeln.de (annexr3-5.slip.Uni-Koeln.DE) by Sisyphos.MI.Uni-Koeln.DE with SMTP id AA27652 (5.67b/IDA-1.5); Wed, 19 Feb 1997 20:59:18 +0100 Received: (from se@localhost) by x14.mi.uni-koeln.de (8.8.5/8.6.9) id UAA02253; Wed, 19 Feb 1997 20:59:15 +0100 (CET) Message-Id: <19970219205914.RB27571@x14.mi.uni-koeln.de> Date: Wed, 19 Feb 1997 20:59:14 +0100 From: se@freebsd.org (Stefan Esser) To: msmith@atrad.adelaide.edu.au (Michael Smith) Cc: se@freebsd.org (Stefan Esser), hardware@freebsd.org Subject: Re: _big_ IDE disks? References: <19970213175603.JR15968@x14.mi.uni-koeln.de> <199702140035.LAA21975@genesis.atrad.adelaide.edu.au> X-Mailer: Mutt 0.60-PL0 Mime-Version: 1.0 In-Reply-To: <199702140035.LAA21975@genesis.atrad.adelaide.edu.au>; from Michael Smith on Feb 14, 1997 11:05:39 +1030 Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Feb 14, msmith@atrad.adelaide.edu.au (Michael Smith) wrote: > wdc0 at 0x1f0-0x1f7 irq 14 flags 0x80ff80ff on isa > wdc0: unit 0 (wd0): , 32-bit, multi-block-16 > wd0: 788MB (10003392 sectors), 9924 cyls, 16 heads, 63 S/T, 512 B/S > IOZONE performance measurements: > 8339742 bytes/second for writing the file > 9316631 bytes/second for reading the file > > -------Sequential Output-------- ---Sequential Input-- --Random-- > -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- > Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU > 128 3858 71.8 8178 28.2 2486 14.1 4210 71.1 8280 28.1 117.0 4.7 Ok. Quite good performance for an (E)IDE drive ... Bruce Evans recently pointed out, that the CPU% load actually is 100%, but the cycles not accounted for are spent in the WD driver's interrupt handler. This indicates, that 30% of a P5-166 is required for 4MB/s, or that the CPU overhead for the PIO mode transfer is some 8% of the P5-166 per MB/s ... The 8MB/s on block transfers should thus correspond to some 65%CPU spent in the interrupt handler, which would mean, that the CPU is near 100% busy doing these transfers (while there would be some 70% of the CPU left for other processes, if the disk was driven by a bus-master controller.) > (Just for amusement, I ran iozone without any of the go-faster options > for the 'wd' driver enabled : > > Writing the 128 Megabyte file, 'iozone.tmp'...22.453125 seconds > Reading the file...23.945312 seconds > > IOZONE performance measurements: > 5977685 bytes/second for writing the file > 5605177 bytes/second for reading the file > > .... so they definitely help 8) We really should be able to compare with a bus-master EIDE driver. The PIIX3 docs (available from Intel's WWW server as a PDF file) give enough detail, to implement DMA transfers for that chip IMHO, so somebody should go for it ... Regards, STefan From owner-freebsd-hardware Wed Feb 19 19:12:04 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA10397 for hardware-outgoing; Wed, 19 Feb 1997 19:12:04 -0800 (PST) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA10213; Wed, 19 Feb 1997 19:09:17 -0800 (PST) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id OAA16623; Thu, 20 Feb 1997 14:07:28 +1100 Date: Thu, 20 Feb 1997 14:07:28 +1100 From: Bruce Evans Message-Id: <199702200307.OAA16623@godzilla.zeta.org.au> To: msmith@atrad.adelaide.edu.au, se@freebsd.org Subject: Re: _big_ IDE disks? Cc: hardware@freebsd.org Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >> wdc0 at 0x1f0-0x1f7 irq 14 flags 0x80ff80ff on isa >> wdc0: unit 0 (wd0): , 32-bit, multi-block-16 >> wd0: 788MB (10003392 sectors), 9924 cyls, 16 heads, 63 S/T, 512 B/S >> -------Sequential Output-------- ---Sequential Input-- --Random-- >> -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- >> Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU >> 128 3858 71.8 8178 28.2 2486 14.1 4210 71.1 8280 28.1 117.0 4.7 > >Ok. Quite good performance for an (E)IDE drive ... I checked the specs for it. It really can do > 8MB/s except possibly on the inner tracks. >Bruce Evans recently pointed out, that the CPU% load actually >is 100%, but the cycles not accounted for are spent in the WD >driver's interrupt handler. This is for the per-char i/o. In PIO4 mode, transferring at 8MB/sec "only" takes 50% of the CPU. It's not quite as inefficient as putc() :-). The 28% overhead for the block mode i/o's is probably mostly for copying from the buffer cache. If the CPU is a P5/166, then i586-optimized copyout is apparently not being used. I think the overhead would be about 20% with it (I see 14% for 5MB/sec on a P5/133). >The 8MB/s on block transfers should thus correspond to some >65%CPU spent in the interrupt handler, which would mean, that Not quite that much. However, 16-bit transfers would take 100% of the CPU. >the CPU is near 100% busy doing these transfers (while there >would be some 70% of the CPU left for other processes, if the >disk was driven by a bus-master controller.) The i/o through the buffer cache will also thrash the L1 and L2 caches. As well as a 20-30% penalty for copying, there will be a delayed penalty for cache misses. This is difficult to measure. I guess it is 10-20%. The bus-mastering controller will also cause some memory access stalls. You will be lucky if it leaves 50% for other processes. >> (Just for amusement, I ran iozone without any of the go-faster options >> for the 'wd' driver enabled : >> >> Writing the 128 Megabyte file, 'iozone.tmp'...22.453125 seconds >> Reading the file...23.945312 seconds >> >> IOZONE performance measurements: >> 5977685 bytes/second for writing the file >> 5605177 bytes/second for reading the file >> >> .... so they definitely help 8) Multi-mode doesn't affect the speed much, except indirectly by reducing the number of interrupts. 16-bit transfers apparently caused saturation before 8MB/sec. >We really should be able to compare with a bus-master EIDE >driver. The PIIX3 docs (available from Intel's WWW server as >a PDF file) give enough detail, to implement DMA transfers >for that chip IMHO, so somebody should go for it ... I compared with the Linux one. The overhead was hard to measure (Linux's hdparm reported confusing (lower) speeds) but seemed to be small. Bruce From owner-freebsd-hardware Wed Feb 19 20:03:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA14278 for hardware-outgoing; Wed, 19 Feb 1997 20:03:50 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA14264; Wed, 19 Feb 1997 20:03:23 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id OAA15277; Thu, 20 Feb 1997 14:30:51 +1030 (CST) From: Michael Smith Message-Id: <199702200400.OAA15277@genesis.atrad.adelaide.edu.au> Subject: Re: _big_ IDE disks? In-Reply-To: <199702200307.OAA16623@godzilla.zeta.org.au> from Bruce Evans at "Feb 20, 97 02:07:28 pm" To: bde@zeta.org.au (Bruce Evans) Date: Thu, 20 Feb 1997 14:30:49 +1030 (CST) Cc: msmith@atrad.adelaide.edu.au, se@freebsd.org, hardware@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Bruce Evans stands accused of saying: > >> wdc0 at 0x1f0-0x1f7 irq 14 flags 0x80ff80ff on isa > >> wdc0: unit 0 (wd0): , 32-bit, multi-block-16 > >> wd0: 788MB (10003392 sectors), 9924 cyls, 16 heads, 63 S/T, 512 B/S > > >> -------Sequential Output-------- ---Sequential Input-- --Random-- > >> -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- > >> Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU > >> 128 3858 71.8 8178 28.2 2486 14.1 4210 71.1 8280 28.1 117.0 4.7 > > > >Ok. Quite good performance for an (E)IDE drive ... > > I checked the specs for it. It really can do > 8MB/s except > possibly on the inner tracks. Yup, this was a basic criteria for it being used instead of a SCSI disk; by the time you compensate for the extra CPU overhead, I figured that it would make a reasonable alternative to a ~3M/sec SCSI disk (which was not available 8( ) > The 28% overhead for the block mode i/o's is probably mostly for copying > from the buffer cache. If the CPU is a P5/166, then i586-optimized > copyout is apparently not being used. I think the overhead would be > about 20% with it (I see 14% for 5MB/sec on a P5/133). How do I go about ascertaining this? npx0 is enabled, and the kernel was build with I586_CPU defined (obviously). The kernel it's running is built with npx.c v 1.31.2.5. > Bruce -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hardware Wed Feb 19 20:38:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA15889 for hardware-outgoing; Wed, 19 Feb 1997 20:38:56 -0800 (PST) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA15876; Wed, 19 Feb 1997 20:38:44 -0800 (PST) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id PAA19813; Thu, 20 Feb 1997 15:37:15 +1100 Date: Thu, 20 Feb 1997 15:37:15 +1100 From: Bruce Evans Message-Id: <199702200437.PAA19813@godzilla.zeta.org.au> To: bde@zeta.org.au, msmith@atrad.adelaide.edu.au Subject: Re: _big_ IDE disks? Cc: hardware@freebsd.org, se@freebsd.org Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >> I checked the specs for it. It really can do > 8MB/s except >> possibly on the inner tracks. > >Yup, this was a basic criteria for it being used instead of a SCSI >disk; by the time you compensate for the extra CPU overhead, I figured >that it would make a reasonable alternative to a ~3M/sec SCSI disk >(which was not available 8( ) The DORS-32160 is 5-6MB/sec, but you only nead a FireballTM to compete with that on transfer speed :-). >> from the buffer cache. If the CPU is a P5/166, then i586-optimized >> copyout is apparently not being used. I think the overhead would be >How do I go about ascertaining this? npx0 is enabled, and the kernel >was build with I586_CPU defined (obviously). The kernel it's running >is built with npx.c v 1.31.2.5. I can't think of a better way than using `cvs log'. It was reenabled in npx.c 1.31.2.6. Other ways: run a debugger and look at the vectors. Run `dd if=/dev/zero of=/dev/null bs=1m count=1000' and complain if the throughtput is much lower than 120MB/sec. Bruce From owner-freebsd-hardware Wed Feb 19 23:41:40 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA26494 for hardware-outgoing; Wed, 19 Feb 1997 23:41:40 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA26489; Wed, 19 Feb 1997 23:41:23 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id SAA17125; Thu, 20 Feb 1997 18:11:10 +1030 (CST) From: Michael Smith Message-Id: <199702200741.SAA17125@genesis.atrad.adelaide.edu.au> Subject: Re: _big_ IDE disks? In-Reply-To: <199702200437.PAA19813@godzilla.zeta.org.au> from Bruce Evans at "Feb 20, 97 03:37:15 pm" To: bde@zeta.org.au (Bruce Evans) Date: Thu, 20 Feb 1997 18:11:08 +1030 (CST) Cc: bde@zeta.org.au, msmith@atrad.adelaide.edu.au, hardware@freebsd.org, se@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Bruce Evans stands accused of saying: > > > >Yup, this was a basic criteria for it being used instead of a SCSI > >disk; by the time you compensate for the extra CPU overhead, I figured > >that it would make a reasonable alternative to a ~3M/sec SCSI disk > >(which was not available 8( ) > > The DORS-32160 is 5-6MB/sec, but you only nead a FireballTM to compete > with that on transfer speed :-). I was contemplating the fireballs, but the image wasn't really what I wanted to pass on to people who expect the machine to be reliable 8) Given your previous nasty remarks about IDE disks and CPU overhead (the machine runs several concurrent compute-heavy jobs) and prior experience with SCSI vs IDE in these systems, I wanted to go as far as possible up the IDE performance curve. > >How do I go about ascertaining this? npx0 is enabled, and the kernel > >was build with I586_CPU defined (obviously). The kernel it's running > >is built with npx.c v 1.31.2.5. > > I can't think of a better way than using `cvs log'. It was reenabled > in npx.c 1.31.2.6. Other ways: run a debugger and look at the vectors. > Run `dd if=/dev/zero of=/dev/null bs=1m count=1000' and complain if > the throughtput is much lower than 120MB/sec. Hmm, 85MB/sec beforehand, 131MB/sec with a new kernel. Thanks for the pointer. New test results : wdc0: unit 0 (wd0): , 32-bit, multi-block-16 wd0: 4884MB (10003392 sectors), 9924 cyls, 16 heads, 63 S/T, 512 B/S Writing the 128 Megabyte file, 'iozone.tmp'...15.578125 seconds Reading the file...14.953125 seconds IOZONE performance measurements: 8615781 bytes/second for writing the file 8975898 bytes/second for reading the file -------Sequential Output-------- ---Sequential Input-- --Random-- -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU 100 3939 72.4 8105 24.5 2026 10.7 4427 71.1 8175 20.3 93.7 3.5 Throughput's not up, but CPU overhead is down. Perhaps we've hit the drive's limit? > Bruce -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hardware Thu Feb 20 04:19:24 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA08973 for hardware-outgoing; Thu, 20 Feb 1997 04:19:24 -0800 (PST) Received: from barrow.uwaterloo.ca (sgkruk@barrow.uwaterloo.ca [129.97.140.32]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA08968 for ; Thu, 20 Feb 1997 04:19:21 -0800 (PST) Received: from localhost (sgkruk@localhost) by barrow.uwaterloo.ca (8.8.5/8.8.5) with SMTP id HAA03239; Thu, 20 Feb 1997 07:18:55 -0500 (EST) Date: Thu, 20 Feb 1997 07:18:55 -0500 (EST) From: "Serge G. Kruk" To: Brett Glass at POST-IW1 cc: agonzalez@starbase.globalpc.net, freebsd-hardware@freebsd.org Subject: Re: Digiboard PC/16e problem In-Reply-To: <88256443.000B4ADC.00@IWND1.infoworld.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 18 Feb 1997, Brett Glass at POST-IW1 wrote: > > For some reason, the FreeBSD drivers don't work with many of the Digi > products. I have an X/em that locks up at the same point. > I have just finished porting a mass of commercial application code from SCO to FreeBSD. With much pleasure I might add. Lastly, to move into the production, I need a 16-port serial card well-supported by FreeBSD. The customer presently has a Digiboard PC/16, which seems half-supported, at best. I will be running 1.6, possibly 1.7 for the forseable future. Remember that the customer depends on this machine. What is my best route? New (from where?) driver for Digiboard cards? New card? If so, which one? Serge From owner-freebsd-hardware Thu Feb 20 06:31:33 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA14913 for hardware-outgoing; Thu, 20 Feb 1997 06:31:33 -0800 (PST) Received: from bacall.lodgenet.com (bacall.lodgenet.com [205.138.147.242]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id GAA14908 for ; Thu, 20 Feb 1997 06:31:29 -0800 (PST) Received: (from mail@localhost) by bacall.lodgenet.com (8.6.12/8.6.12) id IAA01946; Thu, 20 Feb 1997 08:29:24 -0600 Received: from garbo.lodgenet.com(204.124.123.250) by bacall via smap (V1.3) id sma001938; Thu Feb 20 08:29:18 1997 Received: from jake.lodgenet.com (jake.lodgenet.com [10.0.11.30]) by garbo.lodgenet.com (8.6.12/8.6.9) with ESMTP id IAA28420; Thu, 20 Feb 1997 08:30:42 -0600 Received: from jake.lodgenet.com (localhost [127.0.0.1]) by jake.lodgenet.com (8.8.5/8.6.12) with ESMTP id IAA28696; Thu, 20 Feb 1997 08:30:10 -0600 (CST) Message-Id: <199702201430.IAA28696@jake.lodgenet.com> X-Mailer: exmh version 2.0beta 12/23/96 To: "Serge G. Kruk" cc: Brett Glass at POST-IW1 , agonzalez@starbase.globalpc.net, freebsd-hardware@freebsd.org Subject: Re: Digiboard PC/16e problem In-reply-to: Your message of "Thu, 20 Feb 1997 07:18:55 EST." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 20 Feb 1997 08:30:10 -0600 From: "Eric L. Hernes" Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk "Serge G. Kruk" writes: > >What is my best route? >New (from where?) driver for Digiboard cards? >New card? If so, which one? I personally like the Stallion cards (www.stallion.com), the driver is written unofficially by a Stallion employee. Engineering level support has been good. We've got at least two machines running 16 modems off Stallions. Comtrol called me the other day saying that they now had a BSD driver. They were going to ship me a card to eval. Their Rocketport gets pretty good reviews in the trade-rags... (maybe their marketing dept is better than their engineering ;-)) The Cyclades is pretty well supported too. I've only got two 8-ports, so I can't say about the 16's. > >Serge > eric. -- erich@lodgenet.com http://rrnet.com/~erich erich@rrnet.com From owner-freebsd-hardware Thu Feb 20 10:17:57 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA27606 for hardware-outgoing; Thu, 20 Feb 1997 10:17:57 -0800 (PST) Received: from kilgour.nething.com (kilgour.nething.com [204.253.210.65]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA27601 for ; Thu, 20 Feb 1997 10:17:53 -0800 (PST) Received: from randy.nething.com (randy.nething.com [204.253.210.83]) by kilgour.nething.com (8.7.5/8.6.9) with SMTP id MAA01899; Thu, 20 Feb 1997 12:16:33 -0600 (CST) Message-Id: <3.0.1.32.19970220121731.006d9adc@nething.com> X-Sender: rberndt@nething.com X-Mailer: Windows Eudora Light Version 3.0.1 beta 4 (32) Date: Thu, 20 Feb 1997 12:17:33 -0600 To: "Serge G. Kruk" From: Randy Berndt Subject: Re: Digiboard PC/16e problem Cc: freebsd-hardware@FreeBSD.ORG Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hardware@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk I like the Boca/16 board. It has modem control, which the boca/8 does not, and it is fully supported by FreeBSD. But I have had it lock up ports if I used more than 12 or so at a time. The Good News (tm): I bought 2 16's and only use the first 8 on each. I have had ZERO trouble with them since. Also, they are so cheap compared to other multiport boards, buying 2 is a reasonable choice. I am using them in to connect to an older minicomputer, so I am only running a mix of 9600 and 19200 connections. At 07:18 AM 2/20/97 -0500, Serge G. Kruk wrote: >I have just finished porting a mass of commercial application code from >SCO to FreeBSD. With much pleasure I might add. Lastly, to move into the >production, I need a 16-port serial card well-supported by FreeBSD. The >customer presently has a Digiboard PC/16, which seems half-supported, at >best. I will be running 1.6, possibly 1.7 for the forseable future. >Remember that the customer depends on this machine. > >What is my best route? >New (from where?) driver for Digiboard cards? >New card? If so, which one? Randy Berndt ---------------------------------- AOS/VS, FreeBSD, DOS: I'm caught in a maze of twisty little command interpreters, all different. From owner-freebsd-hardware Thu Feb 20 11:55:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA03845 for hardware-outgoing; Thu, 20 Feb 1997 11:55:11 -0800 (PST) Received: from lserver.infoworld.com (lserver.infoworld.com [192.216.48.4]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA03837 for ; Thu, 20 Feb 1997 11:55:03 -0800 (PST) Received: from IWND1.infoworld.com (iwnd1.infoworld.com [192.216.49.131]) by lserver.infoworld.com (8.8.5/8.8.5/GNAC-GW-2.1) with SMTP id LAA07235; Thu, 20 Feb 1997 11:53:40 -0800 (PST) Received: by IWND1.infoworld.com(Lotus SMTP MTA v1.05 (305.3 1-15-1997)) id 88256444.006CE85E ; Thu, 20 Feb 1997 11:49:33 -0700 X-Lotus-FromDomain: IW @ IWP From: "Brett Glass at POST-IW1" To: erich@lodgenet.com, sgkruk@barrow.uwaterloo.ca cc: Brett_Glass_at_POST-IW1@infoworld.com, agonzalez@starbase.globalpc.net, freebsd-hardware@freebsd.org Message-ID: <88256444.006B91DE.00@IWND1.infoworld.com> Date: Thu, 20 Feb 1997 12:30:00 -0700 Subject: Re: Digiboard PC/16e problem Mime-Version: 1.0 Content-type: text/plain; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk New Text Item: Re: Digiboard PC/16e problem I hope that Comtrol DOES publish FreeBSD drivers and get them included in FreeBSD 2.2-R. Last I heard from them, they had a beta-quality driver for their ISA cards only. Management seemed doubtful that it was worth the investment to develop other drivers or even complete development on the one they have. I tried to explain that their target market -- ISPs -- were largely running on free UNIX, whereas the markets on which they expending their efforts (NT and Netware) had lower volumes, fierce competition, and high promotional costs. Don't know if my explanations did any good. (Some companies are so blinded by Microsoft that they're unable to recognize that a much bigger market lies on other platforms.) But we can hope.... --Brett ---------- "Serge G. Kruk" writes: > >What is my best route? >New (from where?) driver for Digiboard cards? >New card? If so, which one? I personally like the Stallion cards (www.stallion.com), the driver is written unofficially by a Stallion employee. Engineering level support has been good. We've got at least two machines running 16 modems off Stallions. Comtrol called me the other day saying that they now had a BSD driver. They were going to ship me a card to eval. Their Rocketport gets pretty good reviews in the trade-rags... (maybe their marketing dept is better than their engineering ;-)) The Cyclades is pretty well supported too. I've only got two 8-ports, so I can't say about the 16's. > >Serge > eric. -- erich@lodgenet.com http://rrnet.com/~erich erich@rrnet.com SMTPOriginator: owner-freebsd-hardware@freefall.freebsd.org From owner-freebsd-hardware Thu Feb 20 16:41:48 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA20159 for hardware-outgoing; Thu, 20 Feb 1997 16:41:48 -0800 (PST) Received: from nexgen.ampr.org (max20-105.HiWAAY.net [208.147.153.105]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA20154 for ; Thu, 20 Feb 1997 16:41:39 -0800 (PST) Received: from nexgen (localhost [127.0.0.1]) by nexgen.ampr.org (8.8.5/8.8.4) with ESMTP id SAA28380; Thu, 20 Feb 1997 18:40:54 -0600 (CST) Message-Id: <199702210040.SAA28380@nexgen.ampr.org> X-Mailer: exmh version 1.6.9 8/22/96 To: Bruce Evans cc: hardware@FreeBSD.org From: dkelly@hiwaay.net Subject: Re: _big_ IDE disks? In-reply-to: Message from Bruce Evans of "Thu, 20 Feb 1997 15:37:15 +1100." <199702200437.PAA19813@godzilla.zeta.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 20 Feb 1997 18:40:51 -0600 Sender: owner-hardware@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk Bruce Evans writes: > > Run `dd if=/dev/zero of=/dev/null bs=1m count=1000' and complain if > the throughtput is much lower than 120MB/sec. nexgen: {366} uname -a FreeBSD nexgen 2.2-GAMMA FreeBSD 2.2-GAMMA #0: Sat Feb 8 07:30:46 CST 1997 dkelly@nexgen.ampr.org:/usr/src/sys/compile/NEXGEN i386 nexgen: {367} dd if=/dev/zero of=/dev/null bs=1m count=1000 1000+0 records in 1000+0 records out 1048576000 bytes transferred in 30.300733 secs (34605631 bytes/sec) System is a "P90" NexGen Nx586, PCI. CPU ID's as a 386, and thats the way my kernel is compiled. Who do I complain to? :-) OTOH, it does "make world" in just over 7 hours with its 64 bit wide bus. About the same as my super cheap AMD 5x86/133-P75 on a 32 bit wide bus. Maybe I'm in the market for a new CPU/MB. What gives 120MB/sec performance? This is the $116 5x86/133 w/ 256k of write-thru cache: PeeCee: {1001} dd if=/dev/zero of=/dev/null bs=1m count=1000 1000+0 records in 1000+0 records out 1048576000 bytes transferred in 24.782880 secs (42310498 bytes/sec) PeeCee: {1002} uname -a FreeBSD PeeCee.tbe.com 3.0-CURRENT FreeBSD 3.0-CURRENT #0: Wed Feb 5 13:06:39 CST 1997 dkelly@PeeCee.tbe.com:/usr/src/sys/compile/PEECEE i386 PeeCee would probably beat NexGen in "make world" if I'd replace PeeCee's mode 0 IDE HD with NexGen's 2G narrow Barracuda. mdsc1 102% hinv 1 100 MHZ IP20 Processor FPU: MIPS R4000 Floating Point Coprocessor Revision: 0.0 CPU: MIPS R4000 Processor Chip Revision: 2.2 Hmdsc1 103% time dd if=/dev/zero of=/dev/null bs=1000k count=1000 1000+0 records in 1000+0 records out 0.0u 9.9s 0:10 97% 0+0k 0+0io 0pf+0w mdsc1 104% expr 1048576000 / 10 104857600 Doggnone! Can't even get 120MB/sec out of a moderately old SGI Indigo! But much closer. Hey, this was sorta fun. Time to dig out the 486dx33 and 386sx16. :-) -- David Kelly N4HHE, dkelly@hiwaay.net ===================================================================== The human mind ordinarily operates at only ten percent of its capacity -- the rest is overhead for the operating system. From owner-freebsd-hardware Thu Feb 20 21:56:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA06500 for hardware-outgoing; Thu, 20 Feb 1997 21:56:11 -0800 (PST) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA06495 for ; Thu, 20 Feb 1997 21:56:08 -0800 (PST) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id QAA29486; Fri, 21 Feb 1997 16:43:08 +1100 Date: Fri, 21 Feb 1997 16:43:08 +1100 From: Bruce Evans Message-Id: <199702210543.QAA29486@godzilla.zeta.org.au> To: bde@zeta.org.au, dkelly@hiwaay.net Subject: Re: _big_ IDE disks? Cc: hardware@FreeBSD.org Sender: owner-hardware@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk >> Run `dd if=/dev/zero of=/dev/null bs=1m count=1000' and complain if >> the throughtput is much lower than 120MB/sec. > >nexgen: {366} uname -a >FreeBSD nexgen 2.2-GAMMA FreeBSD 2.2-GAMMA #0: Sat Feb 8 07:30:46 CST 1997 > dkelly@nexgen.ampr.org:/usr/src/sys/compile/NEXGEN i386 >nexgen: {367} dd if=/dev/zero of=/dev/null bs=1m count=1000 >1000+0 records in >1000+0 records out >1048576000 bytes transferred in 30.300733 secs (34605631 bytes/sec) > >System is a "P90" NexGen Nx586, PCI. CPU ID's as a 386, and thats the way >my kernel is compiled. Who do I complain to? :-) My comment only applies to P5's with a 66MHz memory bus. The speed should be much the same for CPU speeds of 100MHz and larger multiples of 33[.3]MHz. The above dd command essentially copies the same 4K kernel buffer to sequential memory 256 times. . The kernel buffer stays in the L1 cache so reading it is almost free and the speed approaches the maximum main memory write speed which is 133M * 4/3 on my P5/133 (I think it is for a 6-2-2-2 burst cycle). Non-Intel x86's aren't detected very well. I think most of them are super 486's, so they can't possibly sustain 170MB/sec writes. >OTOH, it does "make world" in just over 7 hours with its 64 bit wide bus. >About the same as my super cheap AMD 5x86/133-P75 on a 32 bit wide bus. The 64 bit bus might allow it to do 170MB/sec, but even a P5 writes at half speed unless writes go through the FPU (this is for cache misses, which is the usual case for large writes). On P6's, writes normally go at full speed because they are from the cache, but cache lines are normally automatically preallocated, so there are twice as many memory accesses and the write bandwidth is halved. Preallocation is an optimization for normal writes and a pessimization for large writes. >Maybe I'm in the market for a new CPU/MB. What gives 120MB/sec performance? P5's with a 66MHz memory bus. Bruce From owner-freebsd-hardware Fri Feb 21 00:05:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA12700 for hardware-outgoing; Fri, 21 Feb 1997 00:05:13 -0800 (PST) Received: from foo.primenet.com (ip204.sjc.primenet.com [206.165.96.204]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA12695 for ; Fri, 21 Feb 1997 00:05:10 -0800 (PST) Received: (from bkogawa@localhost) by foo.primenet.com (8.8.2/8.6.12) id AAA22530; Fri, 21 Feb 1997 00:05:49 -0800 (PST) Date: Fri, 21 Feb 1997 00:05:49 -0800 (PST) Message-Id: <199702210805.AAA22530@foo.primenet.com> To: bde@zeta.org.au Subject: Re: _big_ IDE disks? Newsgroups: localhost.freebsd.hardware References: <> <199702210543.QAA29486@godzilla.zeta.org.au> From: "Bryan K. Ogawa" Cc: hardware@freebsd.org, bde@zeta.org.au, dkelly@hiwaay.net X-Newsreader: NN version 6.5.0 #1 (NOV) Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In localhost.freebsd.hardware you write: >>> Run `dd if=/dev/zero of=/dev/null bs=1m count=1000' and complain if >>> the throughtput is much lower than 120MB/sec. [...] >My comment only applies to P5's with a 66MHz memory bus. The speed should >be much the same for CPU speeds of 100MHz and larger multiples of 33[.3]MHz. >The above dd command essentially copies the same 4K kernel buffer to >sequential memory 256 times. . The kernel buffer stays in the L1 cache >so reading it is almost free and the speed approaches the maximum main >memory write speed which is 133M * 4/3 on my P5/133 (I think it is for >a 6-2-2-2 burst cycle). Hm. I can only get about 100MB/seg doing this, max (P5/133). I did it with an X server running--should that matter? {foo} ~ 0:01 ttyp3 > uname -a FreeBSD foo.primenet.com 2.2-ALPHA FreeBSD 2.2-ALPHA #5: Sun Jan 12 19:28:01 PST 1997 bkogawa@foo.primenet.com:/usr/src/sys/compile/ATWO i386 {foo} ~ 0:04 ttyp3 > dd if=/dev/zero of=/dev/null bs=1m count=1000 1000+0 records in 1000+0 records out 1048576000 bytes transferred in 10 secs (104857600 bytes/sec) Perhaps my BIOS is mis-set, but I wouldn't think so...? -- bryan k ogawa http://www.primenet.com/~bkogawa/ From owner-freebsd-hardware Fri Feb 21 00:29:30 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA13630 for hardware-outgoing; Fri, 21 Feb 1997 00:29:30 -0800 (PST) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA13624 for ; Fri, 21 Feb 1997 00:29:26 -0800 (PST) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id TAA01446; Fri, 21 Feb 1997 19:24:31 +1100 Date: Fri, 21 Feb 1997 19:24:31 +1100 From: Bruce Evans Message-Id: <199702210824.TAA01446@godzilla.zeta.org.au> To: bde@zeta.org.au, bkogawa@primenet.com Subject: Re: _big_ IDE disks? Cc: dkelly@hiwaay.net, hardware@freebsd.org Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >Hm. I can only get about 100MB/seg doing this, max (P5/133). I did >it with an X server running--should that matter? No. >{foo} ~ 0:01 ttyp3 > uname -a >FreeBSD foo.primenet.com 2.2-ALPHA FreeBSD 2.2-ALPHA #5: Sun Jan 12 19:28:01 PST > 1997 bkogawa@foo.primenet.com:/usr/src/sys/compile/ATWO i386 >{foo} ~ 0:04 ttyp3 > dd if=/dev/zero of=/dev/null bs=1m count=1000 >1000+0 records in >1000+0 records out >1048576000 bytes transferred in 10 secs (104857600 bytes/sec) > >Perhaps my BIOS is mis-set, but I wouldn't think so...? The fast-i586-copyout option is not enabled in 2.2-GAMMA. I forget what it was in ALPHA. It was either not enabled or buggy (it caused panics). ALPHA is really old. It is also apparently missing the 1996/11/12 change to print the time in milliseconds and the 1996/12/14 change to print the time in microseconds. Bruce From owner-freebsd-hardware Fri Feb 21 09:33:01 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA07278 for hardware-outgoing; Fri, 21 Feb 1997 09:33:01 -0800 (PST) Received: from mexico.brainstorm.eu.org (root@mexico.brainstorm.fr [193.56.58.253]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA07237 for ; Fri, 21 Feb 1997 09:32:17 -0800 (PST) Received: from brasil.brainstorm.eu.org (brasil.brainstorm.fr [193.56.58.33]) by mexico.brainstorm.eu.org (8.8.4/8.8.4) with ESMTP id SAA30024 for ; Fri, 21 Feb 1997 18:32:04 +0100 Received: (from uucp@localhost) by brasil.brainstorm.eu.org (8.8.4/8.6.12) with UUCP id SAA21715 for hardware@FreeBSD.org; Fri, 21 Feb 1997 18:31:42 +0100 Received: (from roberto@localhost) by keltia.freenix.fr (8.8.5/keltia-uucp-2.9) id SAA21761; Fri, 21 Feb 1997 18:26:32 +0100 (CET) Message-ID: <19970221182631.YY64393@keltia.freenix.fr> Date: Fri, 21 Feb 1997 18:26:31 +0100 From: roberto@keltia.freenix.fr (Ollivier Robert) To: hardware@FreeBSD.org Subject: Re: _big_ IDE disks? References: <199702210040.SAA28380@nexgen.ampr.org> X-Mailer: Mutt 0.60,1-3,9 Mime-Version: 1.0 X-Operating-System: FreeBSD 3.0-CURRENT ctm#2999 In-Reply-To: <199702210040.SAA28380@nexgen.ampr.org>; from dkelly@hiwaay.net on Feb 20, 1997 18:40:51 -0600 Sender: owner-hardware@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk According to dkelly@hiwaay.net: > System is a "P90" NexGen Nx586, PCI. CPU ID's as a 386, and thats the way > my kernel is compiled. Who do I complain to? :-) Interesting... my 486DX4/100 gets 43 MB/s... > OTOH, it does "make world" in just over 7 hours with its 64 bit wide bus. > About the same as my super cheap AMD 5x86/133-P75 on a 32 bit wide bus. "Make world" is under 4h30 on my machine. (no profiled libs). /usr/src and /usr/obj disks on different SCSI buses. Async mounts. -- Ollivier ROBERT -=- The daemon is FREE! -=- roberto@keltia.freenix.fr FreeBSD keltia.freenix.fr 3.0-CURRENT #39: Sun Feb 2 22:12:44 CET 1997 From owner-freebsd-hardware Fri Feb 21 10:42:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA12040 for hardware-outgoing; Fri, 21 Feb 1997 10:42:39 -0800 (PST) Received: from robkaos.ruhr.de (robkaos.ruhr.de [141.39.227.87]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA12021 for ; Fri, 21 Feb 1997 10:42:31 -0800 (PST) Received: by robkaos.ruhr.de (/\oo/\ Smail3.1.29.1 #29.1) id ; Fri, 21 Feb 97 19:41 MET Message-Id: From: robsch@robkaos.ruhr.de (Robert Schien) Subject: Performance problem with P6-200 To: freebsd-hardware@freebsd.org Date: Fri, 21 Feb 1997 19:41:31 +0100 (MET) X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk The following may be somewhat off topic, but I have some problems with my new P6NP5 motherboard from ASUS and a PPro-200 (256k). When I run the C-linpack or other floating point stuff I can't get more than 14 MFLOPS (with double precision). With single precision I get 69 MFLOPS which is fine. Is it notmal that the P6 has such a lousy double precision performance? What can the reason be? The system has 64 MB EDO-RAM. And yes, all caches are enabled. The C-linpack is compiled with cc from -current (version 2.7.2.1). pgcc doesn't help. The C-linpack doesn't use any special library routine except the floor function. It would be nice if someone with a P6-200 could send me some results of FP benchmarks. BTW, Dhrystone 2.1 delivers 432000 dhrystones/second. Is this ok? TIA for your help. Robert From owner-freebsd-hardware Fri Feb 21 12:21:16 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA19579 for hardware-outgoing; Fri, 21 Feb 1997 12:21:16 -0800 (PST) Received: from george.lbl.gov (george.lbl.gov [128.3.196.93]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA19573 for ; Fri, 21 Feb 1997 12:21:14 -0800 (PST) Received: (jin@localhost) by george.lbl.gov (8.6.10/8.6.5) id MAA18042; Fri, 21 Feb 1997 12:16:56 -0800 Date: Fri, 21 Feb 1997 12:16:56 -0800 From: "Jin Guojun[ITG]" Message-Id: <199702212016.MAA18042@george.lbl.gov> To: freebsd-hardware@freebsd.org, robsch@robkaos.ruhr.de Subject: Re: Performance problem with P6-200 Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > The following may be somewhat off topic, but I have some problems with my new > P6NP5 motherboard from ASUS and a PPro-200 (256k). > When I run the C-linpack or other floating point stuff I can't get > more than 14 MFLOPS (with double precision). With single precision > I get 69 MFLOPS which is fine. Is it notmal that the P6 has such a lousy > double precision performance? > What can the reason be? > The system has 64 MB EDO-RAM. And yes, all caches are enabled. > The C-linpack is compiled with cc from -current (version 2.7.2.1). > pgcc doesn't help. > The C-linpack doesn't use any special library routine except the floor > function. > > It would be nice if someone with a P6-200 could send me some results > of FP benchmarks. > > BTW, Dhrystone 2.1 delivers 432000 dhrystones/second. Is this ok? The FLOPS/DFLOPS ratio I got is: 32/28 for Pentium (Triton) 85/35 for Pentium Pro (440FX same as yours) 69/14 is little strange. -Jin From owner-freebsd-hardware Fri Feb 21 19:02:03 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA15934 for hardware-outgoing; Fri, 21 Feb 1997 19:02:03 -0800 (PST) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA15926 for ; Fri, 21 Feb 1997 19:02:00 -0800 (PST) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id NAA25747; Sat, 22 Feb 1997 13:59:10 +1100 Date: Sat, 22 Feb 1997 13:59:10 +1100 From: Bruce Evans Message-Id: <199702220259.NAA25747@godzilla.zeta.org.au> To: freebsd-hardware@freebsd.org, robsch@robkaos.ruhr.de Subject: Re: Performance problem with P6-200 Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >The following may be somewhat off topic, but I have some problems with my new >P6NP5 motherboard from ASUS and a PPro-200 (256k). >When I run the C-linpack or other floating point stuff I can't get >more than 14 MFLOPS (with double precision). With single precision >I get 69 MFLOPS which is fine. Is it notmal that the P6 has such a lousy >double precision performance? Perhaps many memory variables are misaligned. I forget whether alignment is important for P5's or P6's or both. Anyway, the only support for double alignment is gcc's -malign-double, which is not the default and not supported by the linker. Bruce From owner-freebsd-hardware Fri Feb 21 20:27:54 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA19411 for hardware-outgoing; Fri, 21 Feb 1997 20:27:54 -0800 (PST) Received: from scruz.net (nic.scruz.net [165.227.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA19403 for ; Fri, 21 Feb 1997 20:27:52 -0800 (PST) Received: (from speth@localhost) by scruz.net (8.7.3/1.34) id UAA14952; Fri, 21 Feb 1997 20:27:49 -0800 (PST) Message-Id: <199702220427.UAA14952@scruz.net> From: speth@scruz.net (James G. Speth) Date: Fri, 21 Feb 1997 20:27:47 -0800 X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: hardware@freebsd.org Subject: problem with 2.1.7-RELEASE #0 Cc: speth@nic.scruz.net, matthew@nic.scruz.net Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I installed FreeBSD release 2.1.7 yesterday, and I've been dealing with some very strange errors since then. I'm using a Peak 510S single board computer with an on-board Adaptek AIC7880 ultra-wide SCSI controller (BIOS 1.22S2B), and a Quantum VP32170W drive. The first indication of a problem was during the boot sequence. When I have the SCSI controller set for "Initiate Wide Negotiation: On", I get a parity error during the boot probe. When I try disabling parity checking I get some serious data errors during the probe (e.g. "Quaftum"), and a panic when labelling the disk. So, I turned off wide negotiation, and turned on parity checking. This allows me to boot from the floppy and do a normal install with no problems. After the install is complete things get really weird. When I attempt to compile some software that compiles cleanly under an identical system running FreeBSD 2.1.0-RELEASE #4, I get sporadic internal compiler errors. Every now and then it will compile cleanly with no errors, but the rest of the time it gives me errors like these: cc -g -c billing.c assertion "fragP->fr_offset >= 0" failed: file "/usr/src/gnu/usr.bin/as/write.c" , line 529 cc: Internal compiler error: program as got fatal signal 6 *** Error code 1 cc -g -c billing.c cc: Internal compiler error: program cc1 got fatal signal 11 *** Error code 1 cc -g -c utils.c /var/tmp/cc000363.s: Assembler messages: /var/tmp/cc000363.s:6893: Fatal error:Case value 23199812 unexpected at line 380 of file "/usr/src/gnu/usr.bin/as/write.c" cc -g -c billing.c billing.c: In function `undo': billing.c:3880: internal error--unrecognizable insn: (jump_insn 155 153 156 (set (pc) (UnKnown Unknown)) -1 (nil) (nil)) cc: Internal compiler error: program cc1 got fatal signal 6 *** Error code 1 It is very inconsistent, giving different error messages on different attempts, and sometimes giving no error at all. That is what makes me think it is related to the earlier disk problems, or perhaps something in the caching code. Have you seen problems like this, or heard any other reports of problems with this release? Any help you can give me would be appreciated. Thanks. -- Jim Speth scruz-net engineering From owner-freebsd-hardware Fri Feb 21 21:55:20 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA23596 for hardware-outgoing; Fri, 21 Feb 1997 21:55:20 -0800 (PST) Received: from seabass.progroup.com (catfish.progroup.com [206.24.122.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA23582 for ; Fri, 21 Feb 1997 21:54:48 -0800 (PST) Received: from seabass.progroup.com (seabass.progroup.com [206.24.122.1]) by seabass.progroup.com (8.7.5/8.7.3) with SMTP id VAA00314; Fri, 21 Feb 1997 21:51:01 -0800 (PST) Message-ID: <330E8945.773C2448@ProGroup.com> Date: Fri, 21 Feb 1997 21:51:01 -0800 From: Craig Shaver Organization: Productivity Group, Inc. X-Mailer: Mozilla 3.01 (X11; I; FreeBSD 2.1.5-RELEASE i386) MIME-Version: 1.0 To: "James G. Speth" CC: hardware@freebsd.org, speth@nic.scruz.net, matthew@nic.scruz.net Subject: Re: problem with 2.1.7-RELEASE #0 References: <199702220427.UAA14952@scruz.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk James G. Speth wrote: > > I installed FreeBSD release 2.1.7 yesterday, and I've been dealing with some > very strange errors since then. del .... > the install is complete things get really weird. When I attempt to compile > some software that compiles cleanly under an identical system running > FreeBSD 2.1.0-RELEASE #4, I get sporadic internal compiler errors. Every > now and then it will compile cleanly with no errors, but the rest of the > time it gives me errors like these: > > cc -g -c billing.c > assertion "fragP->fr_offset >= 0" failed: file "/usr/src/gnu/usr.bin/as/write.c" > , line 529 > cc: Internal compiler error: program as got fatal signal 6 > *** Error code 1 hmmmmm, yes, I have seen this before .... > > cc -g -c billing.c > cc: Internal compiler error: program cc1 got fatal signal 11 > *** Error code 1 and this .... > > cc -g -c utils.c > /var/tmp/cc000363.s: Assembler messages: > /var/tmp/cc000363.s:6893: Fatal error:Case value 23199812 unexpected at line 380 > of file "/usr/src/gnu/usr.bin/as/write.c" Haven't seen this one -- yet ... > > cc -g -c billing.c > billing.c: In function `undo': > billing.c:3880: internal error--unrecognizable insn: > (jump_insn 155 153 156 (set (pc) > (UnKnown Unknown)) -1 (nil) > (nil)) > cc: Internal compiler error: program cc1 got fatal signal 6 > *** Error code 1 Yup .... > > It is very inconsistent, giving different error messages on different > attempts, and sometimes giving no error at all. That is what makes me think > it is related to the earlier disk problems, or perhaps something in the > caching code. > > Have you seen problems like this, or heard any other reports of problems with > this release? Any help you can give me would be appreciated. > > Thanks. > > -- > Jim Speth > scruz-net engineering I don't think it is anything wrong with FreeBSD 2.1.7. I just built a new machine for Solaris 2.5.1 x86. It is a Tyan Tomcat III dual pentium. I got the p5-150's and overclocked them by using a 3X ratio on a memory bus speed of 66mhz. No problems with the cpus, but I also got 4 pieces of 32mb edo (128MB). The chips were from TI, and I could not run them at 60ns with aggressive settings in the bios for ram related timings. So I took them back and got Mu chips. They worked great at 60ns with default timings, but not at aggressive timings. When I first built the box I tested it by doing simultaneous compiles of several versions of gcc, gdb, perl, etc. With the TI chips at 60ns settings I got the same kind of errors you did. If I set the bios for the 70ns defaults, I did not get the errors. I also swapped in some other memory I had, and the problems disappeared -- at 60ns timing. So, if you have some other memory available, try swapping it around. Try setting your bios to very conservative settings. Disable the cache. etc, etc, .... -- Craig Shaver (craig@progroup.com) (415)390-0654 Productivity Group POB 60458 Sunnyvale, CA 94088 From owner-freebsd-hardware Fri Feb 21 23:24:18 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA26501 for hardware-outgoing; Fri, 21 Feb 1997 23:24:18 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA26496 for ; Fri, 21 Feb 1997 23:24:12 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id RAA29005; Sat, 22 Feb 1997 17:53:56 +1030 (CST) From: Michael Smith Message-Id: <199702220723.RAA29005@genesis.atrad.adelaide.edu.au> Subject: Re: problem with 2.1.7-RELEASE #0 In-Reply-To: <199702220427.UAA14952@scruz.net> from "James G. Speth" at "Feb 21, 97 08:27:47 pm" To: speth@scruz.net (James G. Speth) Date: Sat, 22 Feb 1997 17:53:55 +1030 (CST) Cc: hardware@freebsd.org, speth@nic.scruz.net, matthew@nic.scruz.net X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk James G. Speth stands accused of saying: > I installed FreeBSD release 2.1.7 yesterday, and I've been dealing with some > very strange errors since then. [...errors elided... > Have you seen problems like this, or heard any other reports of problems with > this release? Any help you can give me would be appreciated. Firstly, I hope you're not overclocking this board, or anything else stupid like that. Have you had a previous release up and running on this particular unit? The problems you're describing are classic memory/cache fault symptoms, so I would be more inclined to be pursuing some RAM for a quick swap, preferably something of a different brand. > Jim Speth -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hardware Sat Feb 22 14:04:46 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA22088 for hardware-outgoing; Sat, 22 Feb 1997 14:04:46 -0800 (PST) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA22081 for ; Sat, 22 Feb 1997 14:04:42 -0800 (PST) Received: from narnia.plutotech.com (localhost [127.0.0.1]) by narnia.plutotech.com (8.8.5/8.7.3) with ESMTP id PAA15511; Sat, 22 Feb 1997 15:04:34 -0700 (MST) Message-Id: <199702222204.PAA15511@narnia.plutotech.com> X-Mailer: exmh version 2.0beta 12/23/96 To: speth@scruz.net (James G. Speth) cc: hardware@freebsd.org, speth@nic.scruz.net, matthew@nic.scruz.net Subject: Re: problem with 2.1.7-RELEASE #0 In-reply-to: Your message of "Fri, 21 Feb 1997 20:27:47 PST." <199702220427.UAA14952@scruz.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 22 Feb 1997 15:04:34 -0700 From: "Justin T. Gibbs" Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >I installed FreeBSD release 2.1.7 yesterday, and I've been dealing with some >very strange errors since then. Your account sounds very much like a termination problem. Many changes have been made to the driver since 2.1.0, one of them being to honor the termination settings stored either in a serial eeprom attached to the adapter or left over from a bios. In the past, the termination was always set regardless of cable configuration or user settings. Could it be that your SBC is not properly configured and/or is not setting the termination properly? To find out, boot with the -v flag and note all of the aic7xxx driver messages. You can also see if you BIOS includes SCSI-Select and modify the setting there to match your configuration. Let me know what information you find. >-- >Jim Speth >scruz-net engineering > > -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations ===========================================