From owner-freebsd-smp Sun Oct 20 19:23:14 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id TAA12560 for smp-outgoing; Sun, 20 Oct 1996 19:23:14 -0700 (PDT) Received: from inet1.inetworld.net (inet1.inetworld.net [204.216.57.10]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id TAA12554 for ; Sun, 20 Oct 1996 19:23:12 -0700 (PDT) Received: (from elya@localhost) by inet1.inetworld.net (8.7.5/8.6.12) id TAA07935 for smp@freebsd.org; Sun, 20 Oct 1996 19:23:11 -0700 (PDT) Date: Sun, 20 Oct 1996 19:23:11 -0700 (PDT) From: Elya Kurktchi Message-Id: <199610210223.TAA07935@inet1.inetworld.net> To: smp@freebsd.org Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk subscribe elya@inetworld.net From owner-freebsd-smp Sun Oct 20 23:08:44 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id XAA26158 for smp-outgoing; Sun, 20 Oct 1996 23:08:44 -0700 (PDT) Received: from fletch.fix.net (root@fletch.fix.net [206.190.71.1]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id XAA26152 for ; Sun, 20 Oct 1996 23:08:41 -0700 (PDT) Received: from dialup.fix.net (lts3-099.snlo.dialup.fix.net [206.190.71.99]) by fletch.fix.net (8.7.6/8.7.3) with SMTP id XAA18258 for ; Sun, 20 Oct 1996 23:09:08 -0700 (PDT) Message-Id: <2.2.32.19961021060838.00674a00@fix.net> X-Sender: bsoben@fix.net (Unverified) X-Mailer: Windows Eudora Pro Version 2.2 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Sun, 20 Oct 1996 23:08:38 -0700 To: smp@freebsd.org From: Barry Soben Subject: subscribe smp Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk subscribe smp From owner-freebsd-smp Mon Oct 21 08:51:55 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id IAA19873 for smp-outgoing; Mon, 21 Oct 1996 08:51:55 -0700 (PDT) Received: from skynet.psc.edu (skynet.psc.edu [128.182.61.116]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id IAA19862 for ; Mon, 21 Oct 1996 08:51:47 -0700 (PDT) Received: from skynet.psc.edu (localhost [127.0.0.1]) by skynet.psc.edu (8.7.4/8.7.4 PSC 1996/04/22 lambert) with ESMTP id LAA07806 for ; Mon, 21 Oct 1996 11:50:59 -0400 (EDT) Message-Id: <199610211550.LAA07806@skynet.psc.edu> To: smp@freebsd.org Date: Mon, 21 Oct 1996 11:50:56 -0400 From: "Joseph C. Lappa" Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk subscribe lappa@psc.edu From owner-freebsd-smp Thu Oct 24 23:23:26 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id XAA25468 for smp-outgoing; Thu, 24 Oct 1996 23:23:26 -0700 (PDT) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id XAA25402; Thu, 24 Oct 1996 23:23:16 -0700 (PDT) Received: from quagmire.ki.net (root@quagmire.ki.net [205.150.102.1]) by who.cdrom.com (8.7.5/8.6.11) with ESMTP id WAA15228 ; Thu, 24 Oct 1996 22:55:26 -0700 (PDT) Received: from localhost (scrappy@localhost) by quagmire.ki.net (8.7.6/8.7.5) with SMTP id BAA23241; Fri, 25 Oct 1996 01:55:24 -0400 (EDT) Date: Fri, 25 Oct 1996 01:55:21 -0400 (EDT) From: "Marc G. Fournier" To: current@freebsd.org cc: smp@freebsd.org Subject: Recommendations... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi... I'm going back to University in the spring, and wish to get together a nice system *before* I become a starving student again :) I'm interested in playing around with the SMP side of FreeBSD, since what I've read on the mail list seems to indicate that its relatively stable, altho still developing... I still haven't looked into costs yet, but would I be better going with a P6 vs a Dual-P5? Does anyone have any recommendations on which motherboard for either I should be looking at? make/model? cache? On a costs note...which would I get more 'bang-for-my-buck' from? Thanks... Marc G. Fournier scrappy@ki.net Systems Administrator @ ki.net scrappy@freebsd.org From owner-freebsd-smp Fri Oct 25 03:55:39 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id DAA08887 for smp-outgoing; Fri, 25 Oct 1996 03:55:39 -0700 (PDT) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id DAA08834; Fri, 25 Oct 1996 03:55:29 -0700 (PDT) Received: from mx.serv.net (mx.serv.net [199.201.191.10]) by who.cdrom.com (8.7.5/8.6.11) with ESMTP id XAA15363 ; Thu, 24 Oct 1996 23:55:07 -0700 (PDT) Received: from MindBender.serv.net by mx.serv.net (8.7.5/SERV Revision: 2.30) id XAA15526; Thu, 24 Oct 1996 23:55:05 -0700 (PDT) Received: from localhost.HeadCandy.com (michaelv@localhost.HeadCandy.com [127.0.0.1]) by MindBender.serv.net (8.7.5/8.7.3) with SMTP id XAA04986; Thu, 24 Oct 1996 23:54:47 -0700 (PDT) Message-Id: <199610250654.XAA04986@MindBender.serv.net> X-Authentication-Warning: MindBender.serv.net: Host michaelv@localhost.HeadCandy.com [127.0.0.1] didn't use HELO protocol To: "Marc G. Fournier" cc: current@freebsd.org, smp@freebsd.org Subject: Re: Recommendations... In-reply-to: Your message of Fri, 25 Oct 96 01:55:21 -0400. Date: Thu, 24 Oct 1996 23:54:46 -0700 From: "Michael L. VanLoon -- HeadCandy.com" Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I still haven't looked into costs yet, but would I be >better going with a P6 vs a Dual-P5? Does anyone have any >recommendations on which motherboard for either I should be >looking at? make/model? cache? > On a costs note...which would I get more 'bang-for-my-buck' >from? P6's will definitely give you higher performance. But, 200MHz P6s are almost impossible to get right now, and because of that, have gone way back up in price. Don't bother with a 180 -- wrong bus speed (remember, always multiples of 33 1/3). The P6 166MHz with the 512K cache is supposed to be a good chip, if you can get it cheap (and faster than the P6/180-256K). I've been told that the Tyan dual P6 motherboard is priced very well. I would avoid SuperMicro -- I suspect their quality control could use some help. However, even a "lowly" Pentium will make a Really Excellent *BSD machine. So, it might not give you as much absolute punch per dollar, but it might be possible to do dual-P5 for less overall dollars (and you could probably get the chips right away). Definitely try to get 512K cache(s) if you go with a dual-P5. For What It's Worth... ----------------------------------------------------------------------------- Michael L. VanLoon michaelv@MindBender.serv.net --< Free your mind and your machine -- NetBSD free un*x >-- NetBSD working ports: 386+PC, Mac 68k, Amiga, Atari 68k, HP300, Sun3, Sun4/4c/4m, DEC MIPS, DEC Alpha, PC532, VAX, MVME68k, arm32... NetBSD ports in progress: PICA, others... ----------------------------------------------------------------------------- From owner-freebsd-smp Fri Oct 25 04:55:47 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id EAA16694 for smp-outgoing; Fri, 25 Oct 1996 04:55:47 -0700 (PDT) Received: from gargoyle.bazzle.com ([206.103.246.190]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id EAA16688; Fri, 25 Oct 1996 04:55:40 -0700 (PDT) Received: from gargoyle.bazzle.com (localhost [127.0.0.1]) by gargoyle.bazzle.com (8.7.6/8.6.12) with SMTP id HAA27658; Fri, 25 Oct 1996 07:55:29 -0400 (EDT) Date: Fri, 25 Oct 1996 07:55:28 -0400 (EDT) From: "Eric J. Chet" To: "Marc G. Fournier" cc: current@freebsd.org, smp@freebsd.org Subject: Re: Recommendations... In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Fri, 25 Oct 1996, Marc G. Fournier wrote: > > Hi... > > I'm going back to University in the spring, and wish to > get together a nice system *before* I become a starving student > again :) > > I'm interested in playing around with the SMP side of > FreeBSD, since what I've read on the mail list seems to indicate > that its relatively stable, altho still developing... > > I still haven't looked into costs yet, but would I be > better going with a P6 vs a Dual-P5? Does anyone have any > recommendations on which motherboard for either I should be > looking at? make/model? cache? > > On a costs note...which would I get more 'bang-for-my-buck' > from? Hello Dual P6s are not cost effective right now. You can get a dual P5, with adaptec 2490UW SCSI onboard, Triton-II, 512K pipeline burst cache mainboard for $350. It's a GigaByte GA-586DX-512. I believe one of the SMP developers is using this mainboard. Check out http://www.atipa.com for a reseller. This board with two P5-133 would make a nice development system. Peace, Eric J. Chet - ejc@bazzle.com > > Thanks... > > Marc G. Fournier scrappy@ki.net > Systems Administrator @ ki.net scrappy@freebsd.org > From owner-freebsd-smp Fri Oct 25 06:02:59 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id GAA19569 for smp-outgoing; Fri, 25 Oct 1996 06:02:59 -0700 (PDT) Received: from po1.glue.umd.edu (po1.glue.umd.edu [129.2.128.44]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id GAA19539; Fri, 25 Oct 1996 06:02:53 -0700 (PDT) Received: from downlink.eng.umd.edu (downlink.eng.umd.edu [129.2.98.182]) by po1.glue.umd.edu (8.8.2/8.7.3) with ESMTP id JAA19831; Fri, 25 Oct 1996 09:02:49 -0400 (EDT) Received: from localhost (chuckr@localhost) by downlink.eng.umd.edu (8.7.5/8.7.3) with SMTP id JAA06945; Fri, 25 Oct 1996 09:02:49 -0400 (EDT) X-Authentication-Warning: downlink.eng.umd.edu: chuckr owned process doing -bs Date: Fri, 25 Oct 1996 09:02:48 -0400 (EDT) From: Chuck Robey X-Sender: chuckr@downlink.eng.umd.edu To: "Marc G. Fournier" cc: current@freebsd.org, smp@freebsd.org Subject: Re: Recommendations... In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Fri, 25 Oct 1996, Marc G. Fournier wrote: > > Hi... > > I'm going back to University in the spring, and wish to > get together a nice system *before* I become a starving student > again :) > > I'm interested in playing around with the SMP side of > FreeBSD, since what I've read on the mail list seems to indicate > that its relatively stable, altho still developing... > > I still haven't looked into costs yet, but would I be > better going with a P6 vs a Dual-P5? Does anyone have any > recommendations on which motherboard for either I should be > looking at? make/model? cache? Well, the p6/200 chips have gone through the roof in price, so they aren't a reasonable deal right now. I can tell you the p6/166, with 512K cache, are available for about 525 each, which isn't so relatively bad. I think that actually compares pretty well with the price of the p5/166, and the preformance is much higher. I haven't yet got my Tyan Titan Pro (I pick it up today) so I can't really make a recommendation about it yet, sorry. I can tell you that I did look around for a long time, and that was my choice, FWIW. > > On a costs note...which would I get more 'bang-for-my-buck' > from? > > Thanks... > > Marc G. Fournier scrappy@ki.net > Systems Administrator @ ki.net scrappy@freebsd.org > > ----------------------------+----------------------------------------------- Chuck Robey | Interests include any kind of voice or data chuckr@eng.umd.edu | communications topic, C programming, and Unix. 9120 Edmonston Ct #302 | Greenbelt, MD 20770 | I run Journey2 and n3lxx, both FreeBSD (301) 220-2114 | version 2.2 current -- and great FUN! ----------------------------+----------------------------------------------- From owner-freebsd-smp Fri Oct 25 08:15:59 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id IAA25871 for smp-outgoing; Fri, 25 Oct 1996 08:15:59 -0700 (PDT) Received: from horst.bfd.com (horst.bfd.com [204.160.242.10]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id IAA25865; Fri, 25 Oct 1996 08:15:56 -0700 (PDT) Received: from harlie (bastion.bfd.com [204.160.242.2]) by horst.bfd.com (8.7.6/8.7.3) with SMTP id IAA16901; Fri, 25 Oct 1996 08:10:15 -0700 (PDT) Date: Fri, 25 Oct 1996 08:10:15 -0700 (PDT) From: "Eric J. Schwertfeger" X-Sender: ejs@harlie To: "Michael L. VanLoon -- HeadCandy.com" cc: "Marc G. Fournier" , current@FreeBSD.org, smp@FreeBSD.org Subject: Re: Recommendations... In-Reply-To: <199610250654.XAA04986@MindBender.serv.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk On Thu, 24 Oct 1996, Michael L. VanLoon -- HeadCandy.com wrote: > almost impossible to get right now, and because of that, have gone way > back up in price. Don't bother with a 180 -- wrong bus speed > (remember, always multiples of 33 1/3). The P6 166MHz with the 512K > cache is supposed to be a good chip, if you can get it cheap (and > faster than the P6/180-256K). I've always wonderd if the P6/150 (I've got a price of $309 on it) could be overclocked to 166 to get the 33mhz bus speed :-) Given that most Intel CPU's can be overclocked by 10-20% with a low failure rate, it might be worth toying with. Of course, if reliability is more important than price, don't. > machine. So, it might not give you as much absolute punch per dollar, > but it might be possible to do dual-P5 for less overall dollars (and > you could probably get the chips right away). Definitely try to get > 512K cache(s) if you go with a dual-P5. Unless price is critical, which I don't believe to be the case. The Tyan Tomcat II (at $250 for an SMP motherboard based on the HX chipset) has problems if you run dual Pentiums, 512K cache, and speed greater than 120 Mhz. Drop any one of the three, and the problem goes away. The problem may be fixed, I haven't heard, and wouldn't trust anyone to get me a fixed MB even if they had. anyway, neither of these should be major issues in this case, but some other speed freak on a budget may be following this conversation. I've found a FIC Natoma motherboard for $260, and a P6/150 for $309. Yes, I am quite tempted. Only the low memory bandwidth of the Natoma chipset (compared to the Triton HX chipset) has made me hesitate. From owner-freebsd-smp Fri Oct 25 08:30:50 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id IAA26688 for smp-outgoing; Fri, 25 Oct 1996 08:30:50 -0700 (PDT) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id IAA26676; Fri, 25 Oct 1996 08:30:46 -0700 (PDT) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.7.5/8.7.3) id IAA10067; Fri, 25 Oct 1996 08:28:27 -0700 (PDT) From: "Rodney W. Grimes" Message-Id: <199610251528.IAA10067@GndRsh.aac.dev.com> Subject: Re: Recommendations... In-Reply-To: from "Eric J. Schwertfeger" at "Oct 25, 96 08:10:15 am" To: ejs@bfd.com (Eric J. Schwertfeger) Date: Fri, 25 Oct 1996 08:28:27 -0700 (PDT) Cc: michaelv@MindBender.serv.net, scrappy@ki.net, current@FreeBSD.org, smp@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-smp@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > > > On Thu, 24 Oct 1996, Michael L. VanLoon -- HeadCandy.com wrote: > > > almost impossible to get right now, and because of that, have gone way > > back up in price. Don't bother with a 180 -- wrong bus speed > > (remember, always multiples of 33 1/3). The P6 166MHz with the 512K > > cache is supposed to be a good chip, if you can get it cheap (and > > faster than the P6/180-256K). > > I've always wonderd if the P6/150 (I've got a price of $309 on it) could > be overclocked to 166 to get the 33mhz bus speed :-) Given that most > Intel CPU's can be overclocked by 10-20% with a low failure rate, it might > be worth toying with. Of course, if reliability is more important than > price, don't. Though many have reported success with overclocking, realize Intel marked that part a ``150'' for good reason, either it failed on the ATE equipment at 166, or they already had more 166 chips than they needed out of production. Recently they have added some changes to the chips marked ``iPP'' on the bottom of the chip so that they will not work if overclocked, especially on the 75Mhz parts, since yeilds are such that all chips now run atleast at a speed of 100Mhz without any problems. Basically smart (depending on how you want to look at it, that could also be ``dumb'') OEM's and resellers where purchasing 75Mhz chips cheaply and overclocking them to 100. Intel fixed that cost saving factor :-). > > machine. So, it might not give you as much absolute punch per dollar, > > but it might be possible to do dual-P5 for less overall dollars (and > > you could probably get the chips right away). Definitely try to get > > 512K cache(s) if you go with a dual-P5. > > Unless price is critical, which I don't believe to be the case. The Tyan > Tomcat II (at $250 for an SMP motherboard based on the HX chipset) has > problems if you run dual Pentiums, 512K cache, and speed greater than 120 > Mhz. Drop any one of the three, and the problem goes away. The problem > may be fixed, I haven't heard, and wouldn't trust anyone to get me a fixed > MB even if they had. > > anyway, neither of these should be major issues in this case, but some > other speed freak on a budget may be following this conversation. I've > found a FIC Natoma motherboard for $260, and a P6/150 for $309. Yes, I am > quite tempted. Only the low memory bandwidth of the Natoma chipset > (compared to the Triton HX chipset) has made me hesitate. What low memory bandwidth on the Natoma??? That thing smokes when comparied to a 430HX chipset. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation, Inc. Reliable computers for FreeBSD From owner-freebsd-smp Fri Oct 25 09:34:02 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id JAA00765 for smp-outgoing; Fri, 25 Oct 1996 09:34:02 -0700 (PDT) Received: from mx.serv.net (mx.serv.net [199.201.191.10]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id JAA00722; Fri, 25 Oct 1996 09:33:52 -0700 (PDT) Received: from MindBender.serv.net by mx.serv.net (8.7.5/SERV Revision: 2.30) id JAA24070; Fri, 25 Oct 1996 09:33:47 -0700 (PDT) Received: from localhost.HeadCandy.com (michaelv@localhost.HeadCandy.com [127.0.0.1]) by MindBender.serv.net (8.7.5/8.7.3) with SMTP id JAA09094; Fri, 25 Oct 1996 09:33:41 -0700 (PDT) Message-Id: <199610251633.JAA09094@MindBender.serv.net> X-Authentication-Warning: MindBender.serv.net: Host michaelv@localhost.HeadCandy.com [127.0.0.1] didn't use HELO protocol To: "Rodney W. Grimes" cc: ejs@bfd.com (Eric J. Schwertfeger), scrappy@ki.net, current@freebsd.org, smp@freebsd.org Subject: Re: Recommendations... In-reply-to: Your message of Fri, 25 Oct 96 08:28:27 -0700. <199610251528.IAA10067@GndRsh.aac.dev.com> Date: Fri, 25 Oct 1996 09:33:41 -0700 From: "Michael L. VanLoon -- HeadCandy.com" Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk [...] >> quite tempted. Only the low memory bandwidth of the Natoma chipset >> (compared to the Triton HX chipset) has made me hesitate. The low memory bandwidth problem was on the _ORION_, not the Natoma. ----------------------------------------------------------------------------- Michael L. VanLoon michaelv@MindBender.serv.net --< Free your mind and your machine -- NetBSD free un*x >-- NetBSD working ports: 386+PC, Mac 68k, Amiga, Atari 68k, HP300, Sun3, Sun4/4c/4m, DEC MIPS, DEC Alpha, PC532, VAX, MVME68k, arm32... NetBSD ports in progress: PICA, others... ----------------------------------------------------------------------------- From owner-freebsd-smp Fri Oct 25 09:44:30 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id JAA01534 for smp-outgoing; Fri, 25 Oct 1996 09:44:30 -0700 (PDT) Received: from clem.systemsix.com (clem.systemsix.com [198.99.86.131]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id JAA01528 for ; Fri, 25 Oct 1996 09:44:25 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by clem.systemsix.com (8.6.12/8.6.12) with SMTP id KAA15885; Fri, 25 Oct 1996 10:44:01 -0600 Message-Id: <199610251644.KAA15885@clem.systemsix.com> X-Authentication-Warning: clem.systemsix.com: Host localhost didn't use HELO protocol X-Mailer: exmh version 1.6.5 12/11/95 From: Steve Passe To: "Eric J. Chet" cc: "Marc G. Fournier" , smp@freebsd.org Subject: Re: Recommendations... In-reply-to: Your message of "Fri, 25 Oct 1996 07:55:28 EDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 25 Oct 1996 10:44:00 -0600 Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, > Dual P6s are not cost effective right now. You can get >a dual P5, with adaptec 2490UW SCSI onboard, Triton-II, 512K >pipeline burst cache mainboard for $350. It's a GigaByte >GA-586DX-512. I believe one of the SMP developers is using this >mainboard. Check out http://www.atipa.com for a reseller. This >board with two P5-133 would make a nice development system. This would be me. I am quite happy with it, its rock solid. If you go the dual P5 route, this is the board to get. I tried overclocking the 133's to 166, but I evidently have some of the newer CPUs that refuse to overclock...bummer! A single P6-166 would be a faster system, at least currently. A dual P6 would be very nice. But beware, for most things the SMP kernel can actually be slower than using a single CPU. I hope that we will change this situation SOON! -- Steve Passe | powered by smp@csn.net | FreeBSD From owner-freebsd-smp Fri Oct 25 11:29:25 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id LAA07131 for smp-outgoing; Fri, 25 Oct 1996 11:29:25 -0700 (PDT) Received: from pcpsj.pfcs.com (harlan.fred.net [205.252.219.31]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id LAA07106; Fri, 25 Oct 1996 11:29:07 -0700 (PDT) Received: from mumps.pfcs.com (mumps.pfcs.com [192.52.69.11]) by pcpsj.pfcs.com (8.6.12/8.6.9) with SMTP id OAA26157; Fri, 25 Oct 1996 14:28:48 -0400 Received: from localhost by mumps.pfcs.com with SMTP id AA03387 (5.67b/IDA-1.5); Fri, 25 Oct 1996 14:28:47 -0400 To: current@freebsd.org, smp@freebsd.org Subject: PIC+EISA Recommendations? Date: Fri, 25 Oct 1996 14:28:45 -0300 Message-Id: <3385.846268125@mumps.pfcs.com> From: Harlan Stenn Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Since this is a topic of discussion... I'm about to pick up a Photon video card so I can use one of my Sony single-sync monitors with FreeBSD. The 4M PCI Photon card is only $45 more than the 2M ISA Photon card, so I'd rather get the PCI version. I have an 486/90 EISA machine with an AHA 1742, and I'd rather not waste that controller if it's cost-effective to move it to a new machine. So is there a good single or dual P[56] or P{5,6} CPU motherboard that has both PCI and EISA? Or should I just leave my 486/90 box alone and get a straight PCI motherboard? Thanks... H From owner-freebsd-smp Fri Oct 25 12:01:41 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA08634 for smp-outgoing; Fri, 25 Oct 1996 12:01:41 -0700 (PDT) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id MAA08626; Fri, 25 Oct 1996 12:01:37 -0700 (PDT) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA14036; Fri, 25 Oct 1996 11:58:16 -0700 From: Terry Lambert Message-Id: <199610251858.LAA14036@phaeton.artisoft.com> Subject: Re: PIC+EISA Recommendations? To: Harlan.Stenn@pfcs.com (Harlan Stenn) Date: Fri, 25 Oct 1996 11:58:16 -0700 (MST) Cc: current@freebsd.org, smp@freebsd.org In-Reply-To: <3385.846268125@mumps.pfcs.com> from "Harlan Stenn" at Oct 25, 96 02:28:45 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Since this is a topic of discussion... > > I'm about to pick up a Photon video card so I can use one of my Sony > single-sync monitors with FreeBSD. > > The 4M PCI Photon card is only $45 more than the 2M ISA Photon card, so > I'd rather get the PCI version. > > I have an 486/90 EISA machine with an AHA 1742, and I'd rather not waste > that controller if it's cost-effective to move it to a new machine. > > So is there a good single or dual P[56] or P{5,6} CPU motherboard that > has both PCI and EISA? The ASUS PCI/EISA is what Poul or Peter is using (on loan from Walnut Creek, it was the machine Jack Vogel, who did the original SMP work and the FreeBSD 1.1.5.1 SPARC port, was using). It's a decent motherboard. I got the PCI/ISA motherboard because I wanted to address the issues that ISA introduces, and so that there would be a hardware difference taken into account in the low level code base. > Or should I just leave my 486/90 box alone and get a straight PCI > motherboard? You simply *can't* get a "straight PCI motherboard". The closest you can get is an industrial box -- there are several companies selling passive PCI backplane machines, with no ISA components at all. Unfortunately, the 82378 PCI-ISA bridge is generally on the processor card itself, instead of off on a seperate card, so there's no way to tell from software what hardware is there, short of burning it into the BIOS ROM's. 8-(. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-smp Fri Oct 25 12:56:12 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA11172 for smp-outgoing; Fri, 25 Oct 1996 12:56:12 -0700 (PDT) Received: from agora.rdrop.com (root@agora.rdrop.com [199.2.210.241]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id MAA11101; Fri, 25 Oct 1996 12:55:50 -0700 (PDT) Received: from clem.systemsix.com by agora.rdrop.com with smtp (Smail3.1.29.1 #17) id m0vGsLy-00097uC; Fri, 25 Oct 96 12:55 PDT Received: from localhost (localhost [127.0.0.1]) by clem.systemsix.com (8.6.12/8.6.12) with SMTP id NAA16741; Fri, 25 Oct 1996 13:50:18 -0600 Message-Id: <199610251950.NAA16741@clem.systemsix.com> X-Authentication-Warning: clem.systemsix.com: Host localhost didn't use HELO protocol X-Mailer: exmh version 1.6.5 12/11/95 From: Steve Passe To: Terry Lambert cc: Harlan.Stenn@pfcs.com (Harlan Stenn), current@freebsd.org, smp@freebsd.org Subject: Re: PIC+EISA Recommendations? In-reply-to: Your message of "Fri, 25 Oct 1996 11:58:16 PDT." <199610251858.LAA14036@phaeton.artisoft.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 25 Oct 1996 13:50:18 -0600 Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, > > Since this is a topic of discussion... > > ... > > I have an 486/90 EISA machine with an AHA 1742, and I'd rather not waste > > that controller if it's cost-effective to move it to a new machine. > > > > So is there a good single or dual P[56] or P{5,6} CPU motherboard that > > has both PCI and EISA? > > The ASUS PCI/EISA is what Poul or Peter is using (on loan from Walnut > Creek, it was the machine Jack Vogel, who did the original SMP work > and the FreeBSD 1.1.5.1 SPARC port, was using). It's a decent > motherboard. > ... We haven't been able to get Peter's MB working with my symmetric IO code and his EISA (2742?) controller. We anticipate other problems with EISA MBs and the MP spec. Quoting Peter: On some EISA systems, some irq's (such as the timer) are hardwired internally to the 8259's and are not available as APIC inputs. This is a problem that will complicate things no end. On those systems both the 8259's and the APICs must be active, with the 8259 feeding it's output through the ExINT on the cpu's local apic. But that puts us well over the 32 interrupt source limit, so we are not even thinking about this yet until the basics are working right. [ end quote ] For the above and other reasons I would suggest avoiding an EISA board for SMP work. If you don't anticipate NEEDing to use EISA cards don't get an EISA MB. This view may change later when we know more about them in an SMP environment, but that is likely to lag other development. -- Steve Passe | powered by smp@csn.net | FreeBSD From owner-freebsd-smp Fri Oct 25 15:23:44 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id PAA25373 for smp-outgoing; Fri, 25 Oct 1996 15:23:44 -0700 (PDT) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id PAA25328 for ; Fri, 25 Oct 1996 15:23:39 -0700 (PDT) Received: from po2.glue.umd.edu (po2.glue.umd.edu [129.2.128.45]) by who.cdrom.com (8.7.5/8.6.11) with ESMTP id OAA17414 for ; Fri, 25 Oct 1996 14:55:54 -0700 (PDT) Received: from carrier.eng.umd.edu (carrier.eng.umd.edu [129.2.98.188]) by po2.glue.umd.edu (8.8.2/8.7.3) with ESMTP id RAA05712 for ; Fri, 25 Oct 1996 17:55:51 -0400 (EDT) Received: from localhost (chuckr@localhost) by carrier.eng.umd.edu (8.7.5/8.7.3) with SMTP id RAA06219 for ; Fri, 25 Oct 1996 17:55:49 -0400 (EDT) X-Authentication-Warning: carrier.eng.umd.edu: chuckr owned process doing -bs Date: Fri, 25 Oct 1996 17:55:48 -0400 (EDT) From: Chuck Robey X-Sender: chuckr@carrier.eng.umd.edu To: freebsd-smp@FreeBSD.ORG Subject: starting smp Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk I now have my new smp board, I want to get current on smp via ctm, the same way I currently do with cvs-cur. I have already downloaded the smp-cur files, but being that ctm is really unhappy about changes made to it's archives, I'm worried about how to start up the smp archive, without ruining my cvs-cur archive in the process. If anyone else is running ctm for cvs-cur and smp-cur, could I get a pointer on how to do it? The smp web page assumes you use sup, which I can't use. Thanks. ----------------------------+----------------------------------------------- Chuck Robey | Interests include any kind of voice or data chuckr@eng.umd.edu | communications topic, C programming, and Unix. 9120 Edmonston Ct #302 | Greenbelt, MD 20770 | I run Journey2 and n3lxx, both FreeBSD (301) 220-2114 | version 2.2 current -- and great FUN! ----------------------------+----------------------------------------------- From owner-freebsd-smp Fri Oct 25 16:01:04 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id QAA29834 for smp-outgoing; Fri, 25 Oct 1996 16:01:04 -0700 (PDT) Received: from sunrise.cs.berkeley.edu (root@sunrise.CS.Berkeley.EDU [128.32.38.121]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id QAA29798; Fri, 25 Oct 1996 16:00:52 -0700 (PDT) Received: (from asami@localhost) by sunrise.cs.berkeley.edu (8.7.5/8.6.12) id PAA29409; Fri, 25 Oct 1996 15:55:08 -0700 (PDT) Date: Fri, 25 Oct 1996 15:55:08 -0700 (PDT) Message-Id: <199610252255.PAA29409@sunrise.cs.berkeley.edu> To: rgrimes@GndRsh.aac.dev.com CC: ejs@bfd.com, michaelv@MindBender.serv.net, scrappy@ki.net, current@FreeBSD.org, smp@FreeBSD.org In-reply-to: <199610251528.IAA10067@GndRsh.aac.dev.com> (rgrimes@gndrsh.aac.dev.com) Subject: Re: Recommendations... From: asami@FreeBSD.org (Satoshi Asami) Sender: owner-smp@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk * What low memory bandwidth on the Natoma??? That thing smokes when comparied * to a 430HX chipset. That contradicts our findings. A P5-133 with Triton or Triton II can move 70-80MB/s (depending on EDO or non-EDO), but I can't get more than 45MB/s out of a P6-200 with Natoma/server (at least that's what Intel told us). Satoshi P.S. Details on "http://now.cs.berkeley.edu/Td/bcopy.html". From owner-freebsd-smp Fri Oct 25 16:11:28 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id QAA01021 for smp-outgoing; Fri, 25 Oct 1996 16:11:28 -0700 (PDT) Received: from clem.systemsix.com (clem.systemsix.com [198.99.86.131]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id QAA01011 for ; Fri, 25 Oct 1996 16:11:23 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by clem.systemsix.com (8.6.12/8.6.12) with SMTP id RAA00389; Fri, 25 Oct 1996 17:10:53 -0600 Message-Id: <199610252310.RAA00389@clem.systemsix.com> X-Authentication-Warning: clem.systemsix.com: Host localhost didn't use HELO protocol X-Mailer: exmh version 1.6.5 12/11/95 From: Steve Passe To: Chuck Robey cc: freebsd-smp@FreeBSD.org Subject: Re: starting smp In-reply-to: Your message of "Fri, 25 Oct 1996 17:55:48 EDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 25 Oct 1996 17:10:53 -0600 Sender: owner-smp@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk Hi, > I now have my new smp board, I want to get current on smp via ctm, the > same way I currently do with cvs-cur. I have already downloaded the > smp-cur files, but being that ctm is really unhappy about changes made to > it's archives, I'm worried about how to start up the smp archive, without > ruining my cvs-cur archive in the process. > > If anyone else is running ctm for cvs-cur and smp-cur, could I get a > pointer on how to do it? The smp web page assumes you use sup, which I > can't use. cc: to smp@csn.net and I'll add this info to the SMP page -- Steve Passe | powered by smp@csn.net | FreeBSD From owner-freebsd-smp Fri Oct 25 16:14:04 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id QAA01271 for smp-outgoing; Fri, 25 Oct 1996 16:14:04 -0700 (PDT) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id QAA01261; Fri, 25 Oct 1996 16:14:01 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by root.com (8.7.6/8.6.5) with SMTP id QAA09230; Fri, 25 Oct 1996 16:15:02 -0700 (PDT) Message-Id: <199610252315.QAA09230@root.com> X-Authentication-Warning: implode.root.com: Host localhost [127.0.0.1] didn't use HELO protocol To: asami@FreeBSD.org (Satoshi Asami) cc: rgrimes@GndRsh.aac.dev.com, ejs@bfd.com, michaelv@MindBender.serv.net, scrappy@ki.net, current@FreeBSD.org, smp@FreeBSD.org Subject: Re: Recommendations... In-reply-to: Your message of "Fri, 25 Oct 1996 15:55:08 PDT." <199610252255.PAA29409@sunrise.cs.berkeley.edu> From: David Greenman Reply-To: dg@Root.COM Date: Fri, 25 Oct 1996 16:15:01 -0700 Sender: owner-smp@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk >P.S. Details on "http://now.cs.berkeley.edu/Td/bcopy.html". BTW, the "fast strings" copy capability of the P6 is often disabled due to it being broken in early rev chips. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-smp Fri Oct 25 17:54:11 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id RAA10496 for smp-outgoing; Fri, 25 Oct 1996 17:54:11 -0700 (PDT) Received: from vinyl.quickweb.com (vinyl.sentex.ca [206.222.77.8]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id RAA10481; Fri, 25 Oct 1996 17:54:07 -0700 (PDT) Received: from localhost (mark@localhost) by vinyl.quickweb.com (8.7.5/8.7.3) with SMTP id UAA16049; Fri, 25 Oct 1996 20:52:38 -0400 (EDT) X-Authentication-Warning: vinyl.quickweb.com: mark owned process doing -bs Date: Fri, 25 Oct 1996 20:52:38 -0400 (EDT) From: Mark Mayo To: Satoshi Asami cc: rgrimes@GndRsh.aac.dev.com, ejs@bfd.com, michaelv@MindBender.serv.net, scrappy@ki.net, current@FreeBSD.org, smp@FreeBSD.org Subject: Re: Recommendations... In-Reply-To: <199610252255.PAA29409@sunrise.cs.berkeley.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk On Fri, 25 Oct 1996, Satoshi Asami wrote: > * What low memory bandwidth on the Natoma??? That thing smokes when comparied > * to a 430HX chipset. > > That contradicts our findings. A P5-133 with Triton or Triton II can > move 70-80MB/s (depending on EDO or non-EDO), but I can't get more > than 45MB/s out of a P6-200 with Natoma/server (at least that's what > Intel told us). That's odd, here are my speeds on a P6-200 with Natoma (440fx)/server board straight from intel: Function Rate (MB/s) RMS time Min time Max time Copy: 76.1639 0.0633 0.0630 0.0648 Scale: 75.5894 0.0636 0.0635 0.0638 Add: 81.3670 0.0886 0.0885 0.0887 Triad: 80.6036 0.0894 0.0893 0.0896 And that was with 12 users logged in and the load at .23. I'd imagine it would be a little faster (~85 MB/s perhaps) in single user mode. It's using 60ns parity RAM. Even my workstation I'm sitting on (a first generation Pro 150 from Digital, 450GX server chipset - funky!!) gets about 65 MB/s average on the STREAM tests. If you're PPro is only doing 45 MB/s, it's time to bring it back and get a new one!! -mark ------------------------------------------- | Mark Mayo mark@quickweb.com | | C-Soft www.quickweb.com | ------------------------------------------- "To iterate is human, to recurse divine." - L. Peter Deutsch > > Satoshi > > P.S. Details on "http://now.cs.berkeley.edu/Td/bcopy.html". > From owner-freebsd-smp Fri Oct 25 18:06:26 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id SAA11148 for smp-outgoing; Fri, 25 Oct 1996 18:06:26 -0700 (PDT) Received: from mail.webspan.net (mail.webspan.net [206.154.70.7]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id SAA11126; Fri, 25 Oct 1996 18:06:20 -0700 (PDT) Received: from orion.webspan.net (orion.webspan.net [206.154.70.5]) by mail.webspan.net (8.7.5/8.7.3) with ESMTP id VAA28315; Fri, 25 Oct 1996 21:05:46 -0400 (EDT) Received: from orion.webspan.net (localhost [127.0.0.1]) by orion.webspan.net (8.7.5/8.7.3) with ESMTP id VAA16724; Fri, 25 Oct 1996 21:04:06 -0400 (EDT) To: Chuck Robey cc: "Marc G. Fournier" , current@freebsd.org, smp@freebsd.org From: "Gary Palmer" Subject: Re: Recommendations... In-reply-to: Your message of "Fri, 25 Oct 1996 09:02:48 EDT." Date: Fri, 25 Oct 1996 21:04:06 -0400 Message-ID: <16722.846291846@orion.webspan.net> Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Chuck Robey wrote in message ID : > I haven't yet got my Tyan Titan Pro (I pick it up today) so I can't really > make a recommendation about it yet, sorry. I can tell you that I did look > around for a long time, and that was my choice, FWIW. I have 2 Tital Pro's at work. Seem like nice motherboards. Up to 1Gb RAM (if you can get the mystical 128Mb SIMMS). Our news box runs on one (yeah, yeah, P6 is overkill, but it makes expire nice and fast :) ) Gary -- Gary Palmer FreeBSD Core Team Member FreeBSD: Turning PC's into workstations. See http://www.FreeBSD.ORG/ for info From owner-freebsd-smp Fri Oct 25 18:43:20 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id SAA14300 for smp-outgoing; Fri, 25 Oct 1996 18:43:20 -0700 (PDT) Received: from po2.glue.umd.edu (po2.glue.umd.edu [129.2.128.45]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id SAA14266; Fri, 25 Oct 1996 18:43:12 -0700 (PDT) Received: from packet.eng.umd.edu (packet.eng.umd.edu [129.2.98.184]) by po2.glue.umd.edu (8.8.2/8.7.3) with ESMTP id VAA12389; Fri, 25 Oct 1996 21:43:09 -0400 (EDT) Received: from localhost (chuckr@localhost) by packet.eng.umd.edu (8.7.5/8.7.3) with SMTP id VAA02604; Fri, 25 Oct 1996 21:43:08 -0400 (EDT) X-Authentication-Warning: packet.eng.umd.edu: chuckr owned process doing -bs Date: Fri, 25 Oct 1996 21:43:08 -0400 (EDT) From: Chuck Robey X-Sender: chuckr@packet.eng.umd.edu To: Gary Palmer cc: "Marc G. Fournier" , current@freebsd.org, smp@freebsd.org Subject: Re: Recommendations... In-Reply-To: <16722.846291846@orion.webspan.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Fri, 25 Oct 1996, Gary Palmer wrote: > Chuck Robey wrote in message ID > : > > I haven't yet got my Tyan Titan Pro (I pick it up today) so I can't really > > make a recommendation about it yet, sorry. I can tell you that I did look > > around for a long time, and that was my choice, FWIW. > > I have 2 Tital Pro's at work. Seem like nice motherboards. Up to 1Gb > RAM (if you can get the mystical 128Mb SIMMS). Our news box runs on > one (yeah, yeah, P6 is overkill, but it makes expire nice and fast :) ) The Asus boards, with 440FX chipsets and dual processors, split the processors up in daughterboards, and were relatively much more expensive for what _seemed_ like the same features, so since I have had such good luck so far with my other Tyan board, I went this way. Paid $2021 for the Titan Pro, 2 p6/166s with 512K cache each, and 64 Megs of EDO ram, from Robert Odell, in Glendale, CA. > > Gary > -- > Gary Palmer FreeBSD Core Team Member > FreeBSD: Turning PC's into workstations. See http://www.FreeBSD.ORG/ for info > ----------------------------+----------------------------------------------- Chuck Robey | Interests include any kind of voice or data chuckr@eng.umd.edu | communications topic, C programming, and Unix. 9120 Edmonston Ct #302 | Greenbelt, MD 20770 | I run Journey2 and n3lxx, both FreeBSD (301) 220-2114 | version 2.2 current -- and great FUN! ----------------------------+----------------------------------------------- From owner-freebsd-smp Fri Oct 25 22:53:37 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id WAA29953 for smp-outgoing; Fri, 25 Oct 1996 22:53:37 -0700 (PDT) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id WAA29943; Fri, 25 Oct 1996 22:53:32 -0700 (PDT) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.7.5/8.7.3) id WAA11725; Fri, 25 Oct 1996 22:53:03 -0700 (PDT) From: "Rodney W. Grimes" Message-Id: <199610260553.WAA11725@GndRsh.aac.dev.com> Subject: Re: Recommendations... In-Reply-To: <199610252255.PAA29409@sunrise.cs.berkeley.edu> from Satoshi Asami at "Oct 25, 96 03:55:08 pm" To: asami@FreeBSD.org (Satoshi Asami) Date: Fri, 25 Oct 1996 22:53:03 -0700 (PDT) Cc: ejs@bfd.com, michaelv@MindBender.serv.net, scrappy@ki.net, current@FreeBSD.org, smp@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-smp@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > * What low memory bandwidth on the Natoma??? That thing smokes when comparied > * to a 430HX chipset. > > That contradicts our findings. A P5-133 with Triton or Triton II can > move 70-80MB/s (depending on EDO or non-EDO), but I can't get more > than 45MB/s out of a P6-200 with Natoma/server (at least that's what > Intel told us). Is this an ``Intel'' box? Ie, made by Intel, if so, so doubt it is probably a piece of ahh.. well... you know... > Satoshi > > P.S. Details on "http://now.cs.berkeley.edu/Td/bcopy.html". Links too too too slow right now to read that.... -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation, Inc. Reliable computers for FreeBSD From owner-freebsd-smp Fri Oct 25 22:56:31 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id WAA00226 for smp-outgoing; Fri, 25 Oct 1996 22:56:31 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id WAA00201; Fri, 25 Oct 1996 22:56:19 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.7.6/8.6.9) id PAA14727; Sat, 26 Oct 1996 15:53:20 +1000 Date: Sat, 26 Oct 1996 15:53:20 +1000 From: Bruce Evans Message-Id: <199610260553.PAA14727@godzilla.zeta.org.au> To: asami@freebsd.org, mark@quickweb.com Subject: Re: Recommendations... Cc: current@freebsd.org, ejs@bfd.com, michaelv@MindBender.serv.net, rgrimes@GndRsh.aac.dev.com, scrappy@ki.net, smp@freebsd.org Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >> * What low memory bandwidth on the Natoma??? That thing smokes when comparied >> * to a 430HX chipset. >> >> That contradicts our findings. A P5-133 with Triton or Triton II can >> move 70-80MB/s (depending on EDO or non-EDO), but I can't get more >> than 45MB/s out of a P6-200 with Natoma/server (at least that's what >> Intel told us). > >That's odd, here are my speeds on a P6-200 with Natoma (440fx)/server >board straight from intel: > >Function Rate (MB/s) RMS time Min time Max time >Copy: 76.1639 0.0633 0.0630 0.0648 >Scale: 75.5894 0.0636 0.0635 0.0638 >Add: 81.3670 0.0886 0.0885 0.0887 >Triad: 80.6036 0.0894 0.0893 0.0896 This is because the 4 Rates reported by the STREAM benchmark are scaled by factors of 2, 2, 3 and 3, respectively, and Natoma is very slow :-). On a P5-133 with Triton 1 (ASUS P55TP4XE) with non-EDO RAM (66 MHz memory clock): Function Rate (MB/s) RMS time Min time Max time Copy: 88.7256 0.1446 0.1443 0.1471 Scale: 80.4207 0.1608 0.1592 0.1624 Add: 89.6191 0.2222 0.2142 0.2318 Triad: 88.3433 0.2232 0.2173 0.2318 This is still slow. This machine can copy at > 75MB/s throughput or 150 MB/s on the same scale as the STREAM tests. Getting this throughput involves prefetching the source bytes a few K at a time and then using FP operations to store them (and perforce FP operations to load them). gcc "optimizes" the Copy benchmark to not use FP at all. This is why the more complicated Add an Triad benchmarks can be faster. I guess the more complicated benchmarks would be speeded up to only about 120MB/s by the same method. The full memory bandwidth of 176MB/sec (on this system) isn't quite reachable even for copying because the FPU is too slow (fistpq takes 6 cycles, which is more than the minimum memory cycle time and leaves no time for loop overheads). Bruce From owner-freebsd-smp Fri Oct 25 23:08:44 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id XAA01449 for smp-outgoing; Fri, 25 Oct 1996 23:08:44 -0700 (PDT) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id XAA01444; Fri, 25 Oct 1996 23:08:40 -0700 (PDT) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.7.5/8.7.3) id XAA11759; Fri, 25 Oct 1996 23:08:06 -0700 (PDT) From: "Rodney W. Grimes" Message-Id: <199610260608.XAA11759@GndRsh.aac.dev.com> Subject: Re: Recommendations... In-Reply-To: <199610252255.PAA29409@sunrise.cs.berkeley.edu> from Satoshi Asami at "Oct 25, 96 03:55:08 pm" To: asami@FreeBSD.org (Satoshi Asami) Date: Fri, 25 Oct 1996 23:08:06 -0700 (PDT) Cc: ejs@bfd.com, michaelv@MindBender.serv.net, scrappy@ki.net, current@FreeBSD.org, smp@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-smp@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > * What low memory bandwidth on the Natoma??? That thing smokes when comparied > * to a 430HX chipset. > > That contradicts our findings. A P5-133 with Triton or Triton II can > move 70-80MB/s (depending on EDO or non-EDO), but I can't get more > than 45MB/s out of a P6-200 with Natoma/server (at least that's what > Intel told us). > > Satoshi > > P.S. Details on "http://now.cs.berkeley.edu/Td/bcopy.html". Okay, finally, the routes came back to life. Your ``P6/Natoma/Intel/server'' was probably a pre production pile of shit. I have a P6 system just getting ready to go to burn in with a 4G quantum Atlas. If you have the time tomorrow I will set it up outside the firewall and let you run benchmarks on it all day long. I think you will find that a current AAI production Pentium PRO 200 system running on the ASUS PCI/I-P6NP5 is quite a blinding fast machine. In this configuration it turns make worlds in about 1:38 minutes, and using multiple spindles it has been seen to go as low as 1:22. All times are 2.1.5 stable with the following /etc/make.conf tweaks: CFLAGS= -O -pipe NOPROFILE= true NOMANCOMPRESS= true SHARED= copies My business has me far to overloaded to run benchmarks for you, but your welcome to come on over and play with the box. I can turn parity off and get another 8 to 12% out of the memory system.... or turn ECC on and slow it down 8 to 12%. Since those require BIOS tweaking it can be arranged for me to do a couple of reboots with tweaks. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation, Inc. Reliable computers for FreeBSD From owner-freebsd-smp Sat Oct 26 14:21:39 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id OAA11819 for smp-outgoing; Sat, 26 Oct 1996 14:21:39 -0700 (PDT) Received: from capella.grayphics.com (root@capella.grayphics.com [207.71.216.2]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id OAA11813 for ; Sat, 26 Oct 1996 14:21:37 -0700 (PDT) Received: (from nick@localhost) by capella.grayphics.com (8.7.5/8.6.12) id OAA22764; Sat, 26 Oct 1996 14:21:44 -0700 (PDT) Date: Sat, 26 Oct 1996 14:21:43 -0700 (PDT) From: Nick Esborn To: freebsd-smp@freebsd.org Subject: Current state of SMP support on FreeBSD Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I've been following the list for a few months now. I know that the SMP kernel is alpha level at the moment. I also know how long development projects can take, even when you're not doing anything else. :) I am basically wondering if anyone has any predictions on how long it will be until this part of FreeBSD is release-ready. I want to know how long it looks like I'll be waiting. I have been so happy with FreeBSD that I have no interest in moving to a different platform for SMP, so I will definitely wait. Thanks. Nick Esborn Grayphics http://www.grayphics.com/ From owner-freebsd-smp Sat Oct 26 16:16:50 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id QAA21610 for smp-outgoing; Sat, 26 Oct 1996 16:16:50 -0700 (PDT) Received: from po1.glue.umd.edu (po1.glue.umd.edu [129.2.128.44]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id QAA21605 for ; Sat, 26 Oct 1996 16:16:48 -0700 (PDT) Received: from packet.eng.umd.edu (packet.eng.umd.edu [129.2.98.184]) by po1.glue.umd.edu (8.8.2/8.7.3) with ESMTP id TAA17389; Sat, 26 Oct 1996 19:16:45 -0400 (EDT) Received: from localhost (chuckr@localhost) by packet.eng.umd.edu (8.7.5/8.7.3) with SMTP id TAA03149; Sat, 26 Oct 1996 19:16:44 -0400 (EDT) X-Authentication-Warning: packet.eng.umd.edu: chuckr owned process doing -bs Date: Sat, 26 Oct 1996 19:16:44 -0400 (EDT) From: Chuck Robey X-Sender: chuckr@packet.eng.umd.edu To: Nick Esborn cc: freebsd-smp@freebsd.org Subject: Re: Current state of SMP support on FreeBSD In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sat, 26 Oct 1996, Nick Esborn wrote: > I've been following the list for a few months now. I know that the SMP > kernel is alpha level at the moment. I also know how long development > projects can take, even when you're not doing anything else. :) > > I am basically wondering if anyone has any predictions on how long it > will be until this part of FreeBSD is release-ready. I want to know how > long it looks like I'll be waiting. I have been so happy with FreeBSD > that I have no interest in moving to a different platform for SMP, so > I will definitely wait. Understand, Nick, that everyone doing it isn't trying to meet any schedule, and there is quite a bit of experimentation going on. If it took less than 2 months go get things going, I'd be shocked, but I wouldn't be too shocked if it took a year. Between those limits, it's not terrifically predictable. There's a smallish group that actually has dual processors for experimentation (I'll be doing it soon :-) ), and this slows things too. > > Thanks. > > Nick Esborn > Grayphics > http://www.grayphics.com/ > ----------------------------+----------------------------------------------- Chuck Robey | Interests include any kind of voice or data chuckr@eng.umd.edu | communications topic, C programming, and Unix. 9120 Edmonston Ct #302 | Greenbelt, MD 20770 | I run Journey2 and n3lxx, both FreeBSD (301) 220-2114 | version 2.2 current -- and great FUN! ----------------------------+----------------------------------------------- From owner-freebsd-smp Sat Oct 26 16:25:54 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id QAA22181 for smp-outgoing; Sat, 26 Oct 1996 16:25:54 -0700 (PDT) Received: from capella.grayphics.com (root@capella.grayphics.com [207.71.216.2]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id QAA22174 for ; Sat, 26 Oct 1996 16:25:52 -0700 (PDT) Received: (from nick@localhost) by capella.grayphics.com (8.7.5/8.6.12) id QAA23968; Sat, 26 Oct 1996 16:25:54 -0700 (PDT) Date: Sat, 26 Oct 1996 16:25:54 -0700 (PDT) From: Nick Esborn To: Chuck Robey cc: freebsd-smp@freebsd.org Subject: Re: Current state of SMP support on FreeBSD In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sat, 26 Oct 1996, Chuck Robey wrote: > Understand, Nick, that everyone doing it isn't trying to meet any > schedule, and there is quite a bit of experimentation going on. If it > took less than 2 months go get things going, I'd be shocked, but I > wouldn't be too shocked if it took a year. Between those limits, it's not > terrifically predictable. There's a smallish group that actually has dual > processors for experimentation (I'll be doing it soon :-) ), and this > slows things too. Yeah, I understand. I know nobody's getting paid for their work on FreeBSD, and that donations aren't too common. I don't have two CPUs yet, just the board, and I'll probably need to buy two new, identical CPUs when I'm ready to try SMP anyway, so it's not a big deal. I am very excited about SMP in FBSD though. Thanks for the comments. Nick Grayphics http://www.grayphics.com/ From owner-freebsd-smp Sat Oct 26 16:54:01 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id QAA23461 for smp-outgoing; Sat, 26 Oct 1996 16:54:01 -0700 (PDT) Received: from po1.glue.umd.edu (po1.glue.umd.edu [129.2.128.44]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id QAA23451 for ; Sat, 26 Oct 1996 16:53:58 -0700 (PDT) Received: from packet.eng.umd.edu (packet.eng.umd.edu [129.2.98.184]) by po1.glue.umd.edu (8.8.2/8.7.3) with ESMTP id TAA18049; Sat, 26 Oct 1996 19:53:55 -0400 (EDT) Received: from localhost (chuckr@localhost) by packet.eng.umd.edu (8.7.5/8.7.3) with SMTP id TAA03101; Sat, 26 Oct 1996 19:53:44 -0400 (EDT) X-Authentication-Warning: packet.eng.umd.edu: chuckr owned process doing -bs Date: Sat, 26 Oct 1996 19:53:44 -0400 (EDT) From: Chuck Robey X-Sender: chuckr@packet.eng.umd.edu To: Nick Esborn cc: freebsd-smp@freebsd.org Subject: Re: Current state of SMP support on FreeBSD In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sat, 26 Oct 1996, Nick Esborn wrote: > On Sat, 26 Oct 1996, Chuck Robey wrote: > > > Understand, Nick, that everyone doing it isn't trying to meet any > > schedule, and there is quite a bit of experimentation going on. If it > > took less than 2 months go get things going, I'd be shocked, but I > > wouldn't be too shocked if it took a year. Between those limits, it's not > > terrifically predictable. There's a smallish group that actually has dual > > processors for experimentation (I'll be doing it soon :-) ), and this > > slows things too. > > Yeah, I understand. I know nobody's getting paid for their work on > FreeBSD, and that donations aren't too common. I don't have two CPUs > yet, just the board, and I'll probably need to buy two new, identical CPUs > when I'm ready to try SMP anyway, so it's not a big deal. I am very > excited about SMP in FBSD though. I was thinking when I read this that it's an odd inversion: for the developers, and those following smp development closely, NOW is the exciting time, when things are beginning to work, and discoveries are being made. In a year from now, it won't be nearly so exciting, I think, but perhaps smp will just lead to new avenues. At that time, I guess the excitement will be open to users. ----------------------------+----------------------------------------------- Chuck Robey | Interests include any kind of voice or data chuckr@eng.umd.edu | communications topic, C programming, and Unix. 9120 Edmonston Ct #302 | Greenbelt, MD 20770 | I run Journey2 and n3lxx, both FreeBSD (301) 220-2114 | version 2.2 current -- and great FUN! ----------------------------+-----------------------------------------------