From owner-freebsd-smp Wed Sep 15 12:11:18 1999 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (nat203.199.mpoweredpc.net [142.177.203.199]) by hub.freebsd.org (Postfix) with ESMTP id E9A8814CE8 for ; Wed, 15 Sep 1999 12:10:59 -0700 (PDT) (envelope-from scrappy@hub.org) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.9.3/8.9.1) with ESMTP id QAA27986 for ; Wed, 15 Sep 1999 16:11:03 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Wed, 15 Sep 1999 16:11:02 -0300 (ADT) From: The Hermit Hacker To: freebsd-smp@freebsd.org Subject: Quad-PIII...exists? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi all... We're looking at upgrading one of our servers heree at work form a Sparc to an x86...right now, I'm really trying to push a FreeBSD vs Solaris server, so am trying to configure the server in such a way that I can go either way... ASUS doesn't make a dual-PIII motherboard yet, and IBM has a Quad-PIII server...but IBM is over-priced, IMHO... Any recommendations? Dual-PIII would be enough, but I'd love to have the Quad capabilities for future upgrades at little cost...if possible? Thanks... Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy Systems Administrator @ hub.org primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 12:37:10 1999 Delivered-To: freebsd-smp@freebsd.org Received: from fast.cs.utah.edu (fast.cs.utah.edu [155.99.212.1]) by hub.freebsd.org (Postfix) with ESMTP id D226B14DF8 for ; Wed, 15 Sep 1999 12:36:58 -0700 (PDT) (envelope-from vanmaren@fast.cs.utah.edu) Received: (from vanmaren@localhost) by fast.cs.utah.edu (8.9.1/8.9.1) id NAA09306; Wed, 15 Sep 1999 13:36:57 -0600 (MDT) Date: Wed, 15 Sep 1999 13:36:57 -0600 (MDT) From: Kevin Van Maren Message-Id: <199909151936.NAA09306@fast.cs.utah.edu> To: freebsd-smp@freebsd.org, scrappy@hub.org Subject: Re: Quad-PIII...exists? Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Hi all... > > We're looking at upgrading one of our servers heree at work form a > Sparc to an x86...right now, I'm really trying to push a FreeBSD vs > Solaris server, so am trying to configure the server in such a way that I > can go either way... > > ASUS doesn't make a dual-PIII motherboard yet, and IBM has a > Quad-PIII server...but IBM is over-priced, IMHO... What? How about the P2B-D(S) It should worh with a PII or PIII. PIII processors use the same MBs as the PII. > Any recommendations? Dual-PIII would be enough, but I'd love to > have the Quad capabilities for future upgrades at little cost...if > possible? Quad => Xeon => $$$. If it is going in a server, I'd recommend the Intel L440GX+ motherboard, since it has DUAL PCI busses. FreeBSD 3.2 runs on it just fine (although it doesn't recognize the AGP->PCI bridge for the 2 66MHz PCI slots, they work great). It's not so badly priced, when you consider the dual-channel Adaptec scsi and integrated ethernet (and 2MB video). It's also pretty cool because you can redirect the BIOS display to an ANSI terminal (ie, xterm window running `tip'). That is a nice feature for a server to have (run completly headless). Kevin Van Maren To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 12:46:12 1999 Delivered-To: freebsd-smp@freebsd.org Received: from synthcom.com (beacon.synthcom.com [198.145.98.253]) by hub.freebsd.org (Postfix) with ESMTP id 7D2D214CEA for ; Wed, 15 Sep 1999 12:45:57 -0700 (PDT) (envelope-from neil@beacon.synthcom.com) Received: from beacon.synthcom.com (beacon.synthcom.com [198.145.98.253] (may be forged)) by synthcom.com (8.9.3/8.9.3) with ESMTP id MAA61333 for ; Wed, 15 Sep 1999 12:59:28 -0700 (PDT) (envelope-from neil@beacon.synthcom.com) Date: Wed, 15 Sep 1999 12:59:28 -0700 (PDT) From: Neil Bradley Cc: freebsd-smp@freebsd.org Subject: Re: Quad-PIII...exists? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > ASUS doesn't make a dual-PIII motherboard yet, and IBM has a > Quad-PIII server...but IBM is over-priced, IMHO... For dual PIII, go for the Supermicro. I've found so many marginal hardware problems with Tyan and Asus that I'd never consider them again. I've had excellent luck with all of my Supermicro boards. I don't know if they make quad boards (though they'd be expensive no matter where you get them from), but they make a line of very solid dual CPU motherboards. -->Neil ------------------------------------------------------------------------------- Neil Bradley Synthcom home : http://www.synthcom.com Synthcom Systems, Inc. For over 95% of projects, C++ is the pet rock fad of ICQ # 29402898 programming. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 12:49:17 1999 Delivered-To: freebsd-smp@freebsd.org Received: from gndrsh.dnsmgr.net (GndRsh.dnsmgr.net [198.145.92.4]) by hub.freebsd.org (Postfix) with ESMTP id 4108614FFE for ; Wed, 15 Sep 1999 12:49:14 -0700 (PDT) (envelope-from freebsd@gndrsh.dnsmgr.net) Received: (from freebsd@localhost) by gndrsh.dnsmgr.net (8.9.3/8.9.3) id MAA39775; Wed, 15 Sep 1999 12:48:57 -0700 (PDT) (envelope-from freebsd) From: "Rodney W. Grimes" Message-Id: <199909151948.MAA39775@gndrsh.dnsmgr.net> Subject: Re: Quad-PIII...exists? In-Reply-To: from The Hermit Hacker at "Sep 15, 1999 04:11:02 pm" To: scrappy@hub.org (The Hermit Hacker) Date: Wed, 15 Sep 1999 12:48:57 -0700 (PDT) Cc: freebsd-smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL54 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > Hi all... > > We're looking at upgrading one of our servers heree at work form a > Sparc to an x86...right now, I'm really trying to push a FreeBSD vs > Solaris server, so am trying to configure the server in such a way that I > can go either way... > > ASUS doesn't make a dual-PIII motherboard yet, and IBM has a > Quad-PIII server...but IBM is over-priced, IMHO... The ASUS P2B-D runs PIII's just fine.... > > Any recommendations? Dual-PIII would be enough, but I'd love to > have the Quad capabilities for future upgrades at little cost...if > possible? > > Thanks... > > Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy > Systems Administrator @ hub.org > primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message > -- Rod Grimes - KD7CAX - (RWG25) rgrimes@gndrsh.dnsmgr.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 14:22:29 1999 Delivered-To: freebsd-smp@freebsd.org Received: from atro.pine.nl (atro.pine.nl [212.206.16.2]) by hub.freebsd.org (Postfix) with ESMTP id C634E14CD3 for ; Wed, 15 Sep 1999 14:22:21 -0700 (PDT) (envelope-from mark@pine.nl) Received: from localhost by atro.pine.nl (8.9.3/8.9.3) with ESMTP id XAA20743; Wed, 15 Sep 1999 23:22:17 +0200 (MET DST) Date: Wed, 15 Sep 1999 23:22:17 +0200 (MET DST) From: Mark Lastdrager To: Kevin Van Maren Cc: freebsd-smp@FreeBSD.ORG, scrappy@hub.org Subject: Re: Quad-PIII...exists? In-Reply-To: <199909151936.NAA09306@fast.cs.utah.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org At Wed, 15 Sep 1999, owner-freebsd-smp@FreeBSD.ORG wrote: >Quad => Xeon => $$$. If it is going in a server, I'd recommend >the Intel L440GX+ motherboard, since it has DUAL PCI busses. >FreeBSD 3.2 runs on it just fine (although it doesn't recognize >the AGP->PCI bridge for the 2 66MHz PCI slots, they work great). >It's not so badly priced, when you consider the dual-channel >Adaptec scsi and integrated ethernet (and 2MB video). I'm running 20 machines with Intel N440BX motherboards with dual PII-450 (almost the same as the L440GX) and having trouble with the ethernet. At very heavy load on the network (running Bonnie on a NetApp NFS mount) the fxp drivers gives up and messages like 'fxp0: driver timeout' appear at the console. Reboot is the only solution. The problem only occurs when running in SMP mode. Mark Lastdrager Pine Internet -- email: mark@lastdrager.nl tel. +31-70-3111010 http://www.pine.nl fax. +31-70-3111011 PGP ID 92BB81D1 -- Security news @ http://security.pine.nl Today's excuse: vapors from evaporating sticky-note adhesives To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 16: 1:54 1999 Delivered-To: freebsd-smp@freebsd.org Received: from gndrsh.dnsmgr.net (GndRsh.dnsmgr.net [198.145.92.4]) by hub.freebsd.org (Postfix) with ESMTP id E8D481546F for ; Wed, 15 Sep 1999 16:01:45 -0700 (PDT) (envelope-from freebsd@gndrsh.dnsmgr.net) Received: (from freebsd@localhost) by gndrsh.dnsmgr.net (8.9.3/8.9.3) id PAA41354; Wed, 15 Sep 1999 15:52:31 -0700 (PDT) (envelope-from freebsd) From: "Rodney W. Grimes" Message-Id: <199909152252.PAA41354@gndrsh.dnsmgr.net> Subject: Re: Quad-PIII...exists? In-Reply-To: from Neil Bradley at "Sep 15, 1999 12:59:28 pm" To: neil@synthcom.com (Neil Bradley) Date: Wed, 15 Sep 1999 15:52:30 -0700 (PDT) Cc: freebsd-smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL54 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > ASUS doesn't make a dual-PIII motherboard yet, and IBM has a > > Quad-PIII server...but IBM is over-priced, IMHO... > > For dual PIII, go for the Supermicro. I've found so many marginal hardware > problems with Tyan and Asus that I'd never consider them again. I've had > excellent luck with all of my Supermicro boards. I don't know if they make > quad boards (though they'd be expensive no matter where you get them > from), but they make a line of very solid dual CPU motherboards. Humm... and my experience here is similiar, but in the opposite direction, I won't touch a Supermicro or Tyan board, but ASUS has been rock solid for us, sans occasional problems with BIOS code. And for a ``Tiawanese in-expensive board'' we do a big volume in Soltek, have had outstanding luck with them, and since we are an official distributor we get top nouch service when the rare RMA comes up. They are a little late with thier Dual PII BX board, but given the good luck we have always had with the other products I am going to give a few a ``test drive''. Care to expand on the ``marginal hardware problems''? -- Rod Grimes - KD7CAX - (RWG25) rgrimes@gndrsh.dnsmgr.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 16:19:34 1999 Delivered-To: freebsd-smp@freebsd.org Received: from aurora.sol.net (aurora.sol.net [206.55.65.76]) by hub.freebsd.org (Postfix) with ESMTP id 3264C14F44 for ; Wed, 15 Sep 1999 16:18:47 -0700 (PDT) (envelope-from jgreco@aurora.sol.net) Received: (from jgreco@localhost) by aurora.sol.net (8.9.2/8.9.2/SNNS-1.02) id SAA85092; Wed, 15 Sep 1999 18:18:42 -0500 (CDT) From: Joe Greco Message-Id: <199909152318.SAA85092@aurora.sol.net> Subject: Re: Quad-PIII...exists? To: scrappy@hub.org, smp@freebsd.org Date: Wed, 15 Sep 1999 18:18:42 -0500 (CDT) X-Mailer: ELM [version 2.4ME+ PL43 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > ASUS doesn't make a dual-PIII motherboard yet, and IBM has a > Quad-PIII server...but IBM is over-priced, IMHO... > > Any recommendations? Dual-PIII would be enough, but I'd love to > have the Quad capabilities for future upgrades at little cost...if > possible? What about the ASUS P2B-DS (dual with SCSI), ~~$450USD? (or -D without SCSI) These are supposed to do PIII's. I think there's also an XLG-DS that'll do Slot 2 CPU's, but I haven't seen one yet. I've got at least eight of the P2B-DS's and they work _great_ with PII's, and I don't have any reason to think that they would be different with PIII's. You'll find prices increasing rapidly as you proceed into quad territory. ... Joe ------------------------------------------------------------------------------- Joe Greco - Systems Administrator jgreco@ns.sol.net Solaria Public Access UNIX - Milwaukee, WI 414/342-4847 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 16:28:48 1999 Delivered-To: freebsd-smp@freebsd.org Received: from implode.root.com (root.com [209.102.106.178]) by hub.freebsd.org (Postfix) with ESMTP id D994D1546F for ; Wed, 15 Sep 1999 16:26:34 -0700 (PDT) (envelope-from dg@implode.root.com) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.8/8.8.5) with ESMTP id QAA11720; Wed, 15 Sep 1999 16:25:27 -0700 (PDT) Message-Id: <199909152325.QAA11720@implode.root.com> To: "Rodney W. Grimes" Cc: neil@synthcom.com (Neil Bradley), freebsd-smp@FreeBSD.ORG Subject: Re: Quad-PIII...exists? In-reply-to: Your message of "Wed, 15 Sep 1999 15:52:30 PDT." <199909152252.PAA41354@gndrsh.dnsmgr.net> From: David Greenman Reply-To: dg@root.com Date: Wed, 15 Sep 1999 16:25:27 -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >> > ASUS doesn't make a dual-PIII motherboard yet, and IBM has a >> > Quad-PIII server...but IBM is over-priced, IMHO... >> >> For dual PIII, go for the Supermicro. I've found so many marginal hardware >> problems with Tyan and Asus that I'd never consider them again. I've had >> excellent luck with all of my Supermicro boards. I don't know if they make >> quad boards (though they'd be expensive no matter where you get them >> from), but they make a line of very solid dual CPU motherboards. > >Humm... and my experience here is similiar, but in the opposite direction, >I won't touch a Supermicro or Tyan board, but ASUS has been rock solid >for us, sans occasional problems with BIOS code. ...and while we're all saying conflicting things, I've had *zero* problems with SMP Tyan motherboards (and we're talking sample size >50 motherboards). -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project - http://www.freebsd.org Creator of high-performance Internet servers - http://www.terasolutions.com Pave the road of life with opportunities. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 16:34: 4 1999 Delivered-To: freebsd-smp@freebsd.org Received: from zippy.cdrom.com (zippy.cdrom.com [204.216.27.228]) by hub.freebsd.org (Postfix) with ESMTP id ADF41153FB for ; Wed, 15 Sep 1999 16:33:11 -0700 (PDT) (envelope-from jkh@zippy.cdrom.com) Received: from localhost (jkh@localhost [127.0.0.1]) by zippy.cdrom.com (8.9.3/8.9.3) with ESMTP id QAA90340; Wed, 15 Sep 1999 16:32:25 -0700 (PDT) (envelope-from jkh@zippy.cdrom.com) To: dg@root.com Cc: "Rodney W. Grimes" , neil@synthcom.com (Neil Bradley), freebsd-smp@FreeBSD.ORG Subject: Re: Quad-PIII...exists? In-reply-to: Your message of "Wed, 15 Sep 1999 16:25:27 PDT." <199909152325.QAA11720@implode.root.com> Date: Wed, 15 Sep 1999 16:32:25 -0700 Message-ID: <90336.937438345@localhost> From: "Jordan K. Hubbard" Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > ...and while we're all saying conflicting things, I've had *zero* problems > with SMP Tyan motherboards (and we're talking sample size >50 motherboards). Yes, I have had similarly good luck with my 2 Tyan Thunder 100s. I'm using one to compose this message, and it's been a rock for over a year. No problem with any of the ASUS P2B motherboards at work, either. - Jordan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 16:38:35 1999 Delivered-To: freebsd-smp@freebsd.org Received: from mail.rdc1.ab.home.com (ha1.rdc1.ab.wave.home.com [24.64.2.50]) by hub.freebsd.org (Postfix) with ESMTP id 8519F15369 for ; Wed, 15 Sep 1999 16:38:16 -0700 (PDT) (envelope-from vipw@home.com) Received: from fatman ([24.66.198.169]) by mail.rdc1.ab.home.com (InterMail v4.01.01.07 201-229-111-110) with SMTP id <19990915233815.YNDY7737.mail.rdc1.ab.home.com@fatman> for ; Wed, 15 Sep 1999 16:38:15 -0700 From: "Adam" To: Subject: RE: Quad-PIII...exists? Date: Wed, 15 Sep 1999 17:42:00 -0600 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) In-Reply-To: <90336.937438345@localhost> Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I myself use a Asus P2B-D and I have had no problems with it. It's a very stable motherboard. The Tyan Thunder 100's are also very nice, stay away from TekRam's el'cheapo SMP boards however--I've heard horror stories. -- Adam -----Original Message----- From: owner-freebsd-smp@FreeBSD.ORG [mailto:owner-freebsd-smp@FreeBSD.ORG]On Behalf Of Jordan K. Hubbard Sent: Wednesday, September 15, 1999 5:32 PM To: dg@root.com Cc: Rodney W. Grimes; Neil Bradley; freebsd-smp@FreeBSD.ORG Subject: Re: Quad-PIII...exists? > ...and while we're all saying conflicting things, I've had *zero* problems > with SMP Tyan motherboards (and we're talking sample size >50 motherboards). Yes, I have had similarly good luck with my 2 Tyan Thunder 100s. I'm using one to compose this message, and it's been a rock for over a year. No problem with any of the ASUS P2B motherboards at work, either. - Jordan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 17: 8:37 1999 Delivered-To: freebsd-smp@freebsd.org Received: from gndrsh.dnsmgr.net (GndRsh.dnsmgr.net [198.145.92.4]) by hub.freebsd.org (Postfix) with ESMTP id 3F5D414C86 for ; Wed, 15 Sep 1999 17:08:33 -0700 (PDT) (envelope-from freebsd@gndrsh.dnsmgr.net) Received: (from freebsd@localhost) by gndrsh.dnsmgr.net (8.9.3/8.9.3) id RAA41942; Wed, 15 Sep 1999 17:07:53 -0700 (PDT) (envelope-from freebsd) From: "Rodney W. Grimes" Message-Id: <199909160007.RAA41942@gndrsh.dnsmgr.net> Subject: Re: Quad-PIII...exists? In-Reply-To: <199909152325.QAA11720@implode.root.com> from David Greenman at "Sep 15, 1999 04:25:27 pm" To: dg@root.com Date: Wed, 15 Sep 1999 17:07:52 -0700 (PDT) Cc: neil@synthcom.com (Neil Bradley), freebsd-smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL54 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > >> > ASUS doesn't make a dual-PIII motherboard yet, and IBM has a > >> > Quad-PIII server...but IBM is over-priced, IMHO... > >> > >> For dual PIII, go for the Supermicro. I've found so many marginal hardware > >> problems with Tyan and Asus that I'd never consider them again. I've had > >> excellent luck with all of my Supermicro boards. I don't know if they make > >> quad boards (though they'd be expensive no matter where you get them > >> from), but they make a line of very solid dual CPU motherboards. > > > >Humm... and my experience here is similiar, but in the opposite direction, > >I won't touch a Supermicro or Tyan board, but ASUS has been rock solid > >for us, sans occasional problems with BIOS code. > > ...and while we're all saying conflicting things, I've had *zero* problems > with SMP Tyan motherboards (and we're talking sample size >50 motherboards). You call $10K worth of returned memory 0 problems.... :-) -- Rod Grimes - KD7CAX - (RWG25) rgrimes@gndrsh.dnsmgr.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 18:21: 7 1999 Delivered-To: freebsd-smp@freebsd.org Received: from implode.root.com (root.com [209.102.106.178]) by hub.freebsd.org (Postfix) with ESMTP id 53E1014C36 for ; Wed, 15 Sep 1999 18:21:04 -0700 (PDT) (envelope-from dg@implode.root.com) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.8/8.8.5) with ESMTP id SAA12057; Wed, 15 Sep 1999 18:19:13 -0700 (PDT) Message-Id: <199909160119.SAA12057@implode.root.com> To: "Rodney W. Grimes" Cc: neil@synthcom.com (Neil Bradley), freebsd-smp@FreeBSD.ORG Subject: Re: Quad-PIII...exists? In-reply-to: Your message of "Wed, 15 Sep 1999 17:07:52 PDT." <199909160007.RAA41942@gndrsh.dnsmgr.net> From: David Greenman Reply-To: dg@root.com Date: Wed, 15 Sep 1999 18:19:13 -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >> >> > ASUS doesn't make a dual-PIII motherboard yet, and IBM has a >> >> > Quad-PIII server...but IBM is over-priced, IMHO... >> >> >> >> For dual PIII, go for the Supermicro. I've found so many marginal hardware >> >> problems with Tyan and Asus that I'd never consider them again. I've had >> >> excellent luck with all of my Supermicro boards. I don't know if they make >> >> quad boards (though they'd be expensive no matter where you get them >> >> from), but they make a line of very solid dual CPU motherboards. >> > >> >Humm... and my experience here is similiar, but in the opposite direction, >> >I won't touch a Supermicro or Tyan board, but ASUS has been rock solid >> >for us, sans occasional problems with BIOS code. >> >> ...and while we're all saying conflicting things, I've had *zero* problems >> with SMP Tyan motherboards (and we're talking sample size >50 motherboards). > >You call $10K worth of returned memory 0 problems.... :-) It was the wrong type of memory for that amount of chip loading for that (Intel) chipset. It had nothing to do with the specific motherboard. I've had no problems with those systems or any others that I've built when using the correct memory. -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project - http://www.freebsd.org Creator of high-performance Internet servers - http://www.terasolutions.com Pave the road of life with opportunities. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 15 22:21:30 1999 Delivered-To: freebsd-smp@freebsd.org Received: from gndrsh.dnsmgr.net (GndRsh.dnsmgr.net [198.145.92.4]) by hub.freebsd.org (Postfix) with ESMTP id B9C3314BED for ; Wed, 15 Sep 1999 22:21:24 -0700 (PDT) (envelope-from freebsd@gndrsh.dnsmgr.net) Received: (from freebsd@localhost) by gndrsh.dnsmgr.net (8.9.3/8.9.3) id WAA42461; Wed, 15 Sep 1999 22:20:38 -0700 (PDT) (envelope-from freebsd) From: "Rodney W. Grimes" Message-Id: <199909160520.WAA42461@gndrsh.dnsmgr.net> Subject: Re: Quad-PIII...exists? In-Reply-To: <199909160119.SAA12057@implode.root.com> from David Greenman at "Sep 15, 1999 06:19:13 pm" To: dg@root.com Date: Wed, 15 Sep 1999 22:20:38 -0700 (PDT) Cc: neil@synthcom.com (Neil Bradley), freebsd-smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL54 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > >> >> > ASUS doesn't make a dual-PIII motherboard yet, and IBM has a > >> >> > Quad-PIII server...but IBM is over-priced, IMHO... > >> >> > >> >> For dual PIII, go for the Supermicro. I've found so many marginal hardware > >> >> problems with Tyan and Asus that I'd never consider them again. I've had > >> >> excellent luck with all of my Supermicro boards. I don't know if they make > >> >> quad boards (though they'd be expensive no matter where you get them > >> >> from), but they make a line of very solid dual CPU motherboards. > >> > > >> >Humm... and my experience here is similiar, but in the opposite direction, > >> >I won't touch a Supermicro or Tyan board, but ASUS has been rock solid > >> >for us, sans occasional problems with BIOS code. > >> > >> ...and while we're all saying conflicting things, I've had *zero* problems > >> with SMP Tyan motherboards (and we're talking sample size >50 motherboards). > > > >You call $10K worth of returned memory 0 problems.... :-) > > It was the wrong type of memory for that amount of chip loading for > that (Intel) chipset. It had nothing to do with the specific motherboard. It was a lack of adequate documentation on Tyan's part to not clearly identify that you must have registered dimms when the DIMM size is >256MB. I also seem to recall you saying that even once you got registered DIMMS you still had problems, though not as often, and only when running with 4 slots populated. Not matter how you put it this was a problem, maybe not for you, but certainly for me. I also have problems with a motherboard manufacture who says you must use one of these ``qualified by us'' memory products in our boards or all beats are off. That is design quality through post product design testing, not a good model to follow. Component data sheets exists for a reason. > I've had no problems with those systems or any others that I've built > when using the correct memory. That conflicts with data you have given me in the past. I also will not use any data that says *zero* problems, it means either the sample size is too small (50 is a very small sample in this business) or somehow the data is being filtered so that all the facts are not present. Especially when data to the contrary is avaliable. Also I was not talking about post production product problems, if people like you and myself do our jobs right those should be religated down to the ``field device failure'' class of reliability issues, something we classify completely seperately from ``problems''. Things like having to flash the BIOS because it won't deal with your new fangled PIII chip, but you can't flash the BIOS because it won't boot properly and you don't have an old fangled PII chip just laying around handy, that I consider a ``problem''. Though we really like ASUS for the high end and Soltek for the low end, I admit that we have had _some_ problems with both of them, we have even experienced memory problem very much like the ones you did, though it did not require going to registered DIMMS to solve it. Now where was that setting in the Tyan board to cause AC Power Loss autorestart??? Thats another thing I consider a ``problem'', you a technical oriented person and even you couldn't find it for a long time. Want the funnest one I have had lately with ASUS, well, they went to this ``jumperFree'' design, well, okay, they call it that, but there are still 2 jumpers and 10 DIP switches on the board. Anyway, if you happen to power the system down during POST next time you power it up it drops you into the BIOS setup as it thinks the CPU is set wrong. Well, this doesn't go over very well when you have remote machines on UPS that occasionally have power outages longer than the UPS can hold things up and then the power bounces a few times. I've gone to disabling the jumperFree mode and nail the board settings down. Is this a problem one of our custmers well ever see again, probobably not, is it a problem for someone else buying an ASUS board, well, yes, unless they have happend to read this :-). -- Rod Grimes - KD7CAX - (RWG25) rgrimes@gndrsh.dnsmgr.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 0:39:27 1999 Delivered-To: freebsd-smp@freebsd.org Received: from bofh.dermak.pl (bofh.dermak.pl [212.160.174.2]) by hub.freebsd.org (Postfix) with ESMTP id 4DD1B15856 for ; Thu, 16 Sep 1999 00:39:04 -0700 (PDT) (envelope-from jacke@dermak.pl) Received: from localhost (jacke@localhost) by bofh.dermak.pl (8.9.3/8.9.3) with ESMTP id JAA56243; Thu, 16 Sep 1999 09:37:44 +0200 (CEST) (envelope-from jacke@dermak.pl) Date: Thu, 16 Sep 1999 09:37:42 +0200 (CEST) From: Jakub Klausa To: "Jordan K. Hubbard" Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Quad-PIII...exists? In-Reply-To: <90336.937438345@localhost> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org -----BEGIN PGP SIGNED MESSAGE----- On Wed, 15 Sep 1999, Jordan K. Hubbard wrote: => Yes, I have had similarly good luck with my 2 Tyan Thunder 100s. I'm => using one to compose this message, and it's been a rock for over a year. => No problem with any of the ASUS P2B motherboards at work, either. I've run FreeBSD succesfully on ASUS P2B-DS, Tyan Thunder 100, and now i run it (writing this message from it infact) on Tyan Thunderbolt (GX based). It's all rock solid (Thunderbolt running over a year now, ASUS for some 4 months, and Thunderbolt for 2 weeks without any problem). Just my $0.02 => - Jordan - -- k. -----BEGIN PGP SIGNATURE----- Version: 2.6.3ia Charset: noconv iQCVAwUBN+CeSE54q5P/Kvv1AQFVRAP/adZoJI1oCe2XRvLpzVZb4EWwwCrKBMyi KizzaWQF99PUaLPVVRwDIAH8zfzdQSHQ4B3p2pl7OkiZH7rgx5xQABQCkVWRQkpl Apm0rqWeE8L07wVsZfYKGIN8VqgKyzBQbDT3ok9wMiZune49kyCI8okm6O7lFVb0 t3QsPC0yIK0= =/VFU -----END PGP SIGNATURE----- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 8:28:53 1999 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (nat203.199.mpoweredpc.net [142.177.203.199]) by hub.freebsd.org (Postfix) with ESMTP id 9860E153FA; Thu, 16 Sep 1999 08:28:47 -0700 (PDT) (envelope-from scrappy@hub.org) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.9.3/8.9.1) with ESMTP id MAA36365; Thu, 16 Sep 1999 12:28:29 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Thu, 16 Sep 1999 12:28:29 -0300 (ADT) From: The Hermit Hacker To: bush doctor Cc: Wolfram Schneider , hackers@FreeBSD.ORG, smp@freebsd.org Subject: Re: pmake binaries for Solaris In-Reply-To: <19990916104250.A21719@bantu.cl.msu.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, 16 Sep 1999, bush doctor wrote: > Out of da blue Wolfram Schneider aka (wosch@cs.tu-berlin.de) said: > > I ported pmake to SunOS 5.5.1, 5.6, and 5.7. The port based on the > > FreeBSD-4.0-current make version from 9th September 1999. > > > > I successfully compiled pmake itself and the FreeBSD Web server. > > > > The binaries are available at > > http://www.de.freebsd.org/~wosch/src/pmake/ > This should be nice. The number of solaris boxen I administering > seems to be growing faster that the FBSD boxen. Nice work & thanxs ... I'm still fighting to get a FBSD boxen into the office...have my immediate boss convinced that we should, but her boss is still convinced that Solaris is the only Unix OS out there *sigh* Does anyone have any *good* comparisons between the two? Something like the work that Brad Knowles (?) did recently with Vinum vs DPT SmartCache controllers? If nobody...I'm just working on a Dual PIII upgrade to one of our servers here...if someone could suggest a 'test suite' that I could run on FreeBSD (SMP) before I switched things over to Solaris, I'd be more then willing to run it through so that it was a 'same hardware' scenario. The system (hardware-wise) I'm building is such that it is 'FreeBSD-friendly', as are all the x86 systems I'm building... Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy Systems Administrator @ hub.org primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 9:25:11 1999 Delivered-To: freebsd-smp@freebsd.org Received: from jules.res.cmu.edu (JULES.RES.CMU.EDU [128.2.144.56]) by hub.freebsd.org (Postfix) with ESMTP id 698BB151B9; Thu, 16 Sep 1999 09:25:08 -0700 (PDT) (envelope-from greg@jules.res.cmu.edu) Received: from jules.res.cmu.edu (localhost [127.0.0.1]) by jules.res.cmu.edu (8.9.3/8.9.2) with ESMTP id MAA14962; Thu, 16 Sep 1999 12:25:52 GMT (envelope-from greg@jules.res.cmu.edu) Message-Id: <199909161225.MAA14962@jules.res.cmu.edu> X-Mailer: exmh version 2.0.2 2/24/98 To: freebsd-smp@FreeBSD.ORG, hackers@FreeBSD.ORG Subject: cache-friendly scheduling for SMP Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 16 Sep 1999 12:25:52 +0000 From: greg Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm trying to run 1-2 processes with very large memory footprints on my P2 SMP machine. I'm finding that the process switches cpu's quite often, which obviously isn't good for the caches on the CPU. Can anybody point me to a paper, mailing list discussion, etc. that discusses scheduling processes to not thrash the cpu caches? Or if there's anything in place, how I can take advantage of it, etc. I got stumped on the idea a while ago, so I'm really curious... thanks, Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 11:16:20 1999 Delivered-To: freebsd-smp@freebsd.org Received: from thelma.NVidia.COM (nvgate.nvidia.com [140.174.105.2]) by hub.freebsd.org (Postfix) with ESMTP id EC27B15730 for ; Thu, 16 Sep 1999 11:16:17 -0700 (PDT) (envelope-from dennisw@nvidia.com) Received: from NotesServices.nvidia.com (notes [172.16.30.55]) by thelma.NVidia.COM (8.8.8+Sun/8.8.8) with ESMTP id LAA20491; Thu, 16 Sep 1999 11:12:38 -0700 (PDT) Subject: Re: Quad-PIII...exists? To: "Rodney W. Grimes" Cc: dg@root.com, freebsd-smp@freebsd.org, neil@synthcom.com (Neil Bradley) X-Mailer: Lotus Notes Release 5.0.1 July 16, 1999 Message-ID: From: dennisw@nvidia.com Date: Thu, 16 Sep 1999 11:08:06 -0700 X-Priority: 3 (Normal) X-MIMETrack: Serialize by Router on NotesServices/NVidia/US(Release 5.0.1|July 16, 1999) at 09/16/99 11:15:10 AM MIME-Version: 1.0 Content-type: text/plain; charset=us-ascii Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org So, anyone know of a way to squeeze two dual PIII boards in a mid-sized case? Or know of a company that makes a small case that would house two motherboards? "Rodney W. Grimes" cc: neil@synthcom.com (Neil Bradley), freebsd-smp@freebsd.org Sent by: Subject: Re: Quad-PIII...exists? owner-freebsd-smp@f reebsd.org 09/15/99 10:20 PM > >> >> > ASUS doesn't make a dual-PIII motherboard yet, and IBM has a > >> >> > Quad-PIII server...but IBM is over-priced, IMHO... > >> >> > >> >> For dual PIII, go for the Supermicro. I've found so many marginal hardware > >> >> problems with Tyan and Asus that I'd never consider them again. I've had > >> >> excellent luck with all of my Supermicro boards. I don't know if they make > >> >> quad boards (though they'd be expensive no matter where you get them > >> >> from), but they make a line of very solid dual CPU motherboards. > >> > > >> >Humm... and my experience here is similiar, but in the opposite direction, > >> >I won't touch a Supermicro or Tyan board, but ASUS has been rock solid > >> >for us, sans occasional problems with BIOS code. > >> > >> ...and while we're all saying conflicting things, I've had *zero* problems > >> with SMP Tyan motherboards (and we're talking sample size >50 motherboards). > > > >You call $10K worth of returned memory 0 problems.... :-) > > It was the wrong type of memory for that amount of chip loading for > that (Intel) chipset. It had nothing to do with the specific motherboard. It was a lack of adequate documentation on Tyan's part to not clearly identify that you must have registered dimms when the DIMM size is >256MB. I also seem to recall you saying that even once you got registered DIMMS you still had problems, though not as often, and only when running with 4 slots populated. Not matter how you put it this was a problem, maybe not for you, but certainly for me. I also have problems with a motherboard manufacture who says you must use one of these ``qualified by us'' memory products in our boards or all beats are off. That is design quality through post product design testing, not a good model to follow. Component data sheets exists for a reason. > I've had no problems with those systems or any others that I've built > when using the correct memory. That conflicts with data you have given me in the past. I also will not use any data that says *zero* problems, it means either the sample size is too small (50 is a very small sample in this business) or somehow the data is being filtered so that all the facts are not present. Especially when data to the contrary is avaliable. Also I was not talking about post production product problems, if people like you and myself do our jobs right those should be religated down to the ``field device failure'' class of reliability issues, something we classify completely seperately from ``problems''. Things like having to flash the BIOS because it won't deal with your new fangled PIII chip, but you can't flash the BIOS because it won't boot properly and you don't have an old fangled PII chip just laying around handy, that I consider a ``problem''. Though we really like ASUS for the high end and Soltek for the low end, I admit that we have had _some_ problems with both of them, we have even experienced memory problem very much like the ones you did, though it did not require going to registered DIMMS to solve it. Now where was that setting in the Tyan board to cause AC Power Loss autorestart??? Thats another thing I consider a ``problem'', you a technical oriented person and even you couldn't find it for a long time. Want the funnest one I have had lately with ASUS, well, they went to this ``jumperFree'' design, well, okay, they call it that, but there are still 2 jumpers and 10 DIP switches on the board. Anyway, if you happen to power the system down during POST next time you power it up it drops you into the BIOS setup as it thinks the CPU is set wrong. Well, this doesn't go over very well when you have remote machines on UPS that occasionally have power outages longer than the UPS can hold things up and then the power bounces a few times. I've gone to disabling the jumperFree mode and nail the board settings down. Is this a problem one of our custmers well ever see again, probobably not, is it a problem for someone else buying an ASUS board, well, yes, unless they have happend to read this :-). -- Rod Grimes - KD7CAX - (RWG25) rgrimes@gndrsh.dnsmgr.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 14:33:31 1999 Delivered-To: freebsd-smp@freebsd.org Received: from aurora.sol.net (aurora.sol.net [206.55.65.76]) by hub.freebsd.org (Postfix) with ESMTP id CB1E115746 for ; Thu, 16 Sep 1999 14:33:25 -0700 (PDT) (envelope-from jgreco@aurora.sol.net) Received: (from jgreco@localhost) by aurora.sol.net (8.9.2/8.9.2/SNNS-1.02) id QAA79182; Thu, 16 Sep 1999 16:32:32 -0500 (CDT) From: Joe Greco Message-Id: <199909162132.QAA79182@aurora.sol.net> Subject: Re: Quad-PIII...exists? In-Reply-To: from "dennisw@nvidia.com" at "Sep 16, 1999 6:16:45 pm" To: dennisw@nvidia.com Date: Thu, 16 Sep 1999 16:32:31 -0500 (CDT) Cc: freebsd@gndrsh.dnsmgr.net, freebsd-smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL43 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org [slightly off the SMP topic, I guess] > Want the funnest one I have had lately with ASUS, well, they went > to this ``jumperFree'' design, well, okay, they call it that, but > there are still 2 jumpers and 10 DIP switches on the board. Anyway, > if you happen to power the system down during POST next time you > power it up it drops you into the BIOS setup as it thinks the CPU > is set wrong. Well, this doesn't go over very well when you have > remote machines on UPS that occasionally have power outages longer > than the UPS can hold things up and then the power bounces a few > times. I've gone to disabling the jumperFree mode and nail the board > settings down. Is this a problem one of our custmers well ever see > again, probobably not, is it a problem for someone else buying > an ASUS board, well, yes, unless they have happend to read this :-). Been there done that already. Unfortunately, the machine also fails to respond to the RESET button... it "resets" but goes catatonic instantly, never bringing back video or anything. P3B-F, PIII-450/512, 384MB PC100 ECC Generic AGP VGA AHA 2940U2W SMC 9334 BDT It appears to work OK for some combinations of memory and CPU, usually slower CPUs with small quantities of memory. This doesn't work very well for me with my watchdog rebooter cards... most of my machines have 384MB- 1GB RAM. Have you seen this? Makes them useless as a server-class board. We just went and cleared out Techdata's entire inventory of old P2B-F's, so we'd have something to build servers with until a workaround was found. ... Joe ------------------------------------------------------------------------------- Joe Greco - Systems Administrator jgreco@ns.sol.net Solaria Public Access UNIX - Milwaukee, WI 414/342-4847 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 16:32: 1 1999 Delivered-To: freebsd-smp@freebsd.org Received: from mail-01.cdsnet.net (mail-01.cdsnet.net [206.107.16.35]) by hub.freebsd.org (Postfix) with SMTP id 435DE14FBC for ; Thu, 16 Sep 1999 16:31:53 -0700 (PDT) (envelope-from mrcpu@internetcds.com) Received: (qmail 10603 invoked from network); 16 Sep 1999 23:31:53 -0000 Received: from schizo.cdsnet.net (204.118.244.32) by mail.cdsnet.net with SMTP; 16 Sep 1999 23:31:53 -0000 Date: Thu, 16 Sep 1999 16:31:16 -0700 (PDT) From: Jaye Mathisen X-Sender: mrcpu@schizo.cdsnet.net To: Mark Lastdrager Cc: Kevin Van Maren , freebsd-smp@FreeBSD.ORG, scrappy@hub.org Subject: Re: Quad-PIII...exists? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Yes, I saw similar results with a recent 3.2-stable kernel, but not with a kernel from around July. Same hardware. On the same box, with a kernel from the end of august, top takes about 20 seconds to display the list. Reboot with an older kernel, and it comes up in 1-2 seconds... Never could find a reason, so I ripped it out and put in single UNI boards for this application, and I'm happy. On Wed, 15 Sep 1999, Mark Lastdrager wrote: > At Wed, 15 Sep 1999, owner-freebsd-smp@FreeBSD.ORG wrote: > > >Quad => Xeon => $$$. If it is going in a server, I'd recommend > >the Intel L440GX+ motherboard, since it has DUAL PCI busses. > >FreeBSD 3.2 runs on it just fine (although it doesn't recognize > >the AGP->PCI bridge for the 2 66MHz PCI slots, they work great). > >It's not so badly priced, when you consider the dual-channel > >Adaptec scsi and integrated ethernet (and 2MB video). > > I'm running 20 machines with Intel N440BX motherboards with dual PII-450 > (almost the same as the L440GX) and having trouble with the ethernet. At > very heavy load on the network (running Bonnie on a NetApp NFS mount) the > fxp drivers gives up and messages like 'fxp0: driver timeout' appear at > the console. Reboot is the only solution. The problem only occurs when > running in SMP mode. > > Mark Lastdrager > Pine Internet > > -- > email: mark@lastdrager.nl tel. +31-70-3111010 > http://www.pine.nl fax. +31-70-3111011 > PGP ID 92BB81D1 -- Security news @ http://security.pine.nl > Today's excuse: vapors from evaporating sticky-note adhesives > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 17:17:55 1999 Delivered-To: freebsd-smp@freebsd.org Received: from annwn.phys.washington.edu (annwn.phys.washington.edu [128.95.93.180]) by hub.freebsd.org (Postfix) with ESMTP id 5209F15207 for ; Thu, 16 Sep 1999 17:17:52 -0700 (PDT) (envelope-from somsky@annwn.phys.washington.edu) Received: (from somsky@localhost) by annwn.phys.washington.edu (8.9.3/8.9.3) id RAA19987; Thu, 16 Sep 1999 17:17:52 -0700 (PDT) (envelope-from somsky) Date: Thu, 16 Sep 1999 17:17:51 -0700 From: "William R. Somsky" To: freebsd-smp@freebsd.org Subject: Examples of FreeBSD SMP success? Message-ID: <19990916171751.A19950@annwn.phys.washington.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.7i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org We're considering getting several Intel-based machines here in the UW Physics department for use as desktop work- and computation- stations, and the question has come up whether we would gain much by getting dual-CPU (say dual 500 MHZ Pentium-III) as opposed to single-CPU systems. The anticipated use where dual-CPUs could help us would probably be users either running a computation-intensive job (eg, Mathematica), while simultaneously doing desktop editing/browsing/mailing/TeXing/etc, or running two computation- intensive jobs. (We don't expect that Mathematica or any user job will be multi threaded.) What I've been asked to find out what the state of FreeBSD SMP support is, and if anyone has any real-world examples of using dual-CPUs under FreeBSD that might be similar to this sort of situation and what the results have been. Being as that I've not tried multi-processing under FreeBSD yet, does anybody have any input I can give to my users? ________________________________________________________________________ Dr. William R. Somsky, Unix Mgr somsky@phys.washington.edu Department of Physics, Box 351560 B432 Physics-Astro Bldg Univ. of Washington, Seattle WA 98195-1560 206/616-2954 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 17:42:48 1999 Delivered-To: freebsd-smp@freebsd.org Received: from thomson.iqm.unicamp.br (thomson.iqm.unicamp.br [143.106.13.161]) by hub.freebsd.org (Postfix) with ESMTP id 7236B1535F for ; Thu, 16 Sep 1999 17:42:41 -0700 (PDT) (envelope-from fabio@thomson.iqm.unicamp.br) Received: (from fabio@localhost) by thomson.iqm.unicamp.br (8.9.3/8.9.2) id VAA06428; Thu, 16 Sep 1999 21:42:34 -0300 (EST) (envelope-from fabio) From: Fabio Cesar Gozzo Message-Id: <199909170042.VAA06428@thomson.iqm.unicamp.br> Subject: Re: Examples of FreeBSD SMP success? In-Reply-To: <19990916171751.A19950@annwn.phys.washington.edu> from "William R. Somsky" at "Sep 16, 1999 5:17:51 pm" To: somsky@annwn.phys.washington.edu (William R. Somsky) Date: Thu, 16 Sep 1999 21:42:34 -0300 (EST) Cc: freebsd-smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL43 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org William R. Somsky wrote: >We're considering getting several Intel-based machines here in >the UW Physics department for use as desktop work- and computation- >stations, and the question has come up whether we would gain much >by getting dual-CPU (say dual 500 MHZ Pentium-III) as opposed to >single-CPU systems. > >The anticipated use where dual-CPUs could help us would probably >be users either running a computation-intensive job (eg, Mathematica), >while simultaneously doing desktop editing/browsing/mailing/TeXing/etc, >or running two computation- intensive jobs. (We don't expect that >Mathematica or any user job will be multi threaded.) > >What I've been asked to find out what the state of FreeBSD SMP >support is, and if anyone has any real-world examples of using >dual-CPUs under FreeBSD that might be similar to this sort of >situation and what the results have been. > >Being as that I've not tried multi-processing under FreeBSD yet, >does anybody have any input I can give to my users? I have been used FreeBSD with SMP machines since pre-3.0 for computational chemistry programs and this system has shown to be a very good choice because: 1) FreeBSD SMP is rock solid. I had jobs that took weeks burnning CPU and eating memory/disk resources without a single problem. 2) The cost/performance for this kind of programs is excelent. You just pay for another processor (ok, the MB is also more expensive) and you almost double the performance. I have compared the time required by two sequential jobs with two jobs in parallel and got a ratio around 1.8. It was a Dual PII 300 Mhz and this ratio can be better now with the 100Mhz bus of the new PIII's. Oh, my programs aren't multithread too. In conclusion, all I have to say is go for it ! Regards, -- ************************************************** Fabio Gozzo fabio@iqm.unicamp.br State University of Campinas UNICAMP Chemistry Institute http://thomson.iqm.unicamp.br ************************************************** To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 17:45:29 1999 Delivered-To: freebsd-smp@freebsd.org Received: from csvax.cs.caltech.edu (csvax.cs.caltech.edu [131.215.131.131]) by hub.freebsd.org (Postfix) with ESMTP id 4CA4F157EA for ; Thu, 16 Sep 1999 17:45:18 -0700 (PDT) (envelope-from mika@varese.cs.caltech.edu) Received: from varese.cs.caltech.edu (varese.cs.caltech.edu [131.215.78.28]) by csvax.cs.caltech.edu (8.9.1/8.9.1) with ESMTP id RAA23133; Thu, 16 Sep 1999 17:45:16 -0700 (PDT) Received: from localhost.cs.caltech.edu (localhost.cs.caltech.edu [127.0.0.1]) by varese.cs.caltech.edu (8.8.7/8.7.3) with SMTP id RAA08427; Thu, 16 Sep 1999 17:45:15 -0700 (PDT) Message-Id: <199909170045.RAA08427@varese.cs.caltech.edu> X-Authentication-Warning: varese.cs.caltech.edu: localhost.cs.caltech.edu [127.0.0.1] didn't use HELO protocol To: "William R. Somsky" Cc: freebsd-smp@freebsd.org Subject: Re: Examples of FreeBSD SMP success? In-reply-to: Your message of "Thu, 16 Sep 1999 17:17:51 PDT." <19990916171751.A19950@annwn.phys.washington.edu> Date: Thu, 16 Sep 1999 17:45:15 -0700 From: Mika Nystrom Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello there, I can report that FreeBSD's SMP works great for computation-intensive jobs. We routinely run several copies of spice(-like) circuit simulators that use several hundred megabytes of RAM each and monopolize one CPU each. The efficiency, as long as you are not doing I/O, seems to be 95+% (little different from using two machines for this). We have been doing this for years on dual processor 200 MHz Pentium Pro machines. The response time of interactive jobs and other jobs that do not fall into the category of entirely computation-intensive has improved a great deal in that time (as FreeBSD's SMP has come a long way in this regard, even if it hasn't advanced so much "under the hood"). Regards, Mika Nystrom Asynchronous Systems Architecture Project Department of Computer Science California Institute of Technology "William R. Somsky" writes: >We're considering getting several Intel-based machines here in >the UW Physics department for use as desktop work- and computation- >stations, and the question has come up whether we would gain much >by getting dual-CPU (say dual 500 MHZ Pentium-III) as opposed to >single-CPU systems. > >The anticipated use where dual-CPUs could help us would probably >be users either running a computation-intensive job (eg, Mathematica), >while simultaneously doing desktop editing/browsing/mailing/TeXing/etc, >or running two computation- intensive jobs. (We don't expect that >Mathematica or any user job will be multi threaded.) > >What I've been asked to find out what the state of FreeBSD SMP >support is, and if anyone has any real-world examples of using >dual-CPUs under FreeBSD that might be similar to this sort of >situation and what the results have been. > >Being as that I've not tried multi-processing under FreeBSD yet, >does anybody have any input I can give to my users? > >________________________________________________________________________ >Dr. William R. Somsky, Unix Mgr somsky@phys.washington.ed >u >Department of Physics, Box 351560 B432 Physics-Astro Bldg >Univ. of Washington, Seattle WA 98195-1560 206/616-2954 > > >To Unsubscribe: send mail to majordomo@FreeBSD.org >with "unsubscribe freebsd-smp" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 17:54:59 1999 Delivered-To: freebsd-smp@freebsd.org Received: from pike.sover.net (pike.sover.net [209.198.87.34]) by hub.freebsd.org (Postfix) with ESMTP id 145D7157CE for ; Thu, 16 Sep 1999 17:54:56 -0700 (PDT) (envelope-from adams@digitalspark.net) Received: from nightfall.digitalspark.net (arc0a324.bf.sover.net [209.198.83.232]) by pike.sover.net (8.9.3/8.9.3) with ESMTP id UAA00777; Thu, 16 Sep 1999 20:54:52 -0400 (EDT) Comments: SoVerNet Verification (on pike.sover.net) nightfall.digitalspark.net from arc0a324.bf.sover.net [209.198.83.232] 209.198.83.232 Thu, 16 Sep 1999 20:54:52 -0400 (EDT) Date: Thu, 16 Sep 1999 20:58:38 +0000 (GMT) From: Adam Strohl To: "William R. Somsky" Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Examples of FreeBSD SMP success? In-Reply-To: <19990916171751.A19950@annwn.phys.washington.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, 16 Sep 1999, William R. Somsky wrote: > The anticipated use where dual-CPUs could help us would probably > be users either running a computation-intensive job (eg, Mathematica), > while simultaneously doing desktop editing/browsing/mailing/TeXing/etc, > or running two computation- intensive jobs. (We don't expect that > Mathematica or any user job will be multi threaded.) > > What I've been asked to find out what the state of FreeBSD SMP > support is, and if anyone has any real-world examples of using > dual-CPUs under FreeBSD that might be similar to this sort of > situation and what the results have been. > > Being as that I've not tried multi-processing under FreeBSD yet, > does anybody have any input I can give to my users? Most of the time I do exactly what you describe, run an intesive CPU app in the background, and do desktop work (netscape, vi, some light compiling) at the same time. It runs beautifully. Heavy compiles (even with my crappy IDE disks) take a little more than half the time when I gmake -j 4. Running two copies of seti@home results in twice the number of blocks being cleared in the same amount of time as 1. Desktop work as described above shows little impact. There are things holding FreeBSD SMP back (ie; the big lock in networking, IIRC) but there are TONS of cases where its speed brings a tear to my eye ;'D At this point my machine is heavily I/O bound when doing large disk intensive operations, however things like switching desktops and browsing the web, etc, still are extremely responsive. This is mostly due to FreeBSD's amazing scheduling/priority system, but its even more so on a dual system. With how cheap dual motherboards are now and celeron 370 CPUs/Socket370 -> Slot 1 bridges are its worth it. - ----( Adam Strohl )------------------------------------------------ - - UNIX Operations/Systems http://www.digitalspark.net - - adams (at) digitalspark.net xxx.xxx.xxxx xxxxx - - ----------------------------------------( DigitalSpark.NET )------- - To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 18:41:27 1999 Delivered-To: freebsd-smp@freebsd.org Received: from systemics.com (menger.ai [209.88.68.41]) by hub.freebsd.org (Postfix) with ESMTP id AAD8914CA9 for ; Thu, 16 Sep 1999 18:40:56 -0700 (PDT) (envelope-from iang@systemics.com) Received: (from iang@localhost) by systemics.com (8.9.1/8.9.1) id VAA05868; Thu, 16 Sep 1999 21:40:29 -0400 (AST) (envelope-from iang) Date: Thu, 16 Sep 1999 21:40:29 -0400 (AST) From: Ian Grigg Message-Id: <199909170140.VAA05868@systemics.com> To: freebsd-smp@FreeBSD.ORG, somsky@annwn.phys.washington.edu Subject: Re: Examples of FreeBSD SMP success? Reply-To: iang@systemics.com In-Reply-To: <19990916171751.A19950@annwn.phys.washington.edu> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Back in the 80's I once worked on a job to install a biggish unix box into a Quango [1]. It fairly quickly became apparent that: 1. multi-cpus were more expensive than singles, when adjusted for overall performance. 2. the cost of the project exceeded the cost of the hardware. Under those conditions, even thought the ticket price of the singles was better, I justified to the bean counters that multi was superior because they wouldn't have to pay for another replacement project in 1 year's time [2]. We bought a Sequent, which effectively projected a 5 year useful life rather than 2 years, thus saving 1.5 project costs. I.e., multi-cpus are more expensive, unless you face some bureaucratic cost as well, in which case they will pay for themselves over the extended project life. That was then. Now, you can get dual for a bit more than the price of single [3]. So, assuming 2 cpus, factor 1 is no longer relevant. I've been specifying and buying duals for about a year now, and about half of them are running with dual cpus [4]. I've got the peace of mind knowing that for the other half, they'll still be good machines in 2 years [5]. You're mad if you don't buy duals, even if you don't use the extra slot. There is more time spent in getting them twining, and there *are* some instabilities, as mentioned on the group from time to time. But, 2x500 is way better than 1x500 in a year's time when all the niggles are forgotten, performance is at 1.9, but your budget doesn't support any more machines. iang [1] Quasi-autonomous national government agency. [2] project time was one year! [3] following comments don't apply to quads and up. [4] duals run FreeBSD and Linux. singles are m$. [5] All are ASUS P2B-DS, but I'll probably switch over to Abit P6B duals next time. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 16 22: 3:47 1999 Delivered-To: freebsd-smp@freebsd.org Received: from mail.rdc1.sfba.home.com (ha1.rdc1.sfba.home.com [24.0.0.66]) by hub.freebsd.org (Postfix) with ESMTP id 547F615893; Thu, 16 Sep 1999 22:03:43 -0700 (PDT) (envelope-from adsharma@c62443-a.frmt1.sfba.home.com) Received: from c62443-a.frmt1.sfba.home.com ([24.0.69.165]) by mail.rdc1.sfba.home.com (InterMail v4.01.01.00 201-229-111) with ESMTP id <19990917050343.EHXE29487.mail.rdc1.sfba.home.com@c62443-a.frmt1.sfba.home.com>; Thu, 16 Sep 1999 22:03:43 -0700 Received: (from adsharma@localhost) by c62443-a.frmt1.sfba.home.com (8.9.3/8.9.3) id WAA07157; Thu, 16 Sep 1999 22:03:43 -0700 Date: Thu, 16 Sep 1999 22:03:42 -0700 From: Arun Sharma To: greg Cc: freebsd-smp@FreeBSD.ORG, hackers@FreeBSD.ORG Subject: Re: cache-friendly scheduling for SMP Message-ID: <19990916220342.A7136@home.com> References: <199909161225.MAA14962@jules.res.cmu.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.6i In-Reply-To: <199909161225.MAA14962@jules.res.cmu.edu>; from greg on Thu, Sep 16, 1999 at 12:25:52PM +0000 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, Sep 16, 1999 at 12:25:52PM +0000, greg wrote: > Can anybody point me to a paper, mailing list discussion, etc. that discusses > scheduling processes to not thrash the cpu caches? Or if there's anything in > place, how I can take advantage of it, etc. I got stumped on the idea > a while ago, so I'm really curious... In -current, there is already code to do trivial CPU affinity. Basically, given multiple processes in the same priority queue to choose from, the scheduler will pick the one that last ran on the same CPU. -Arun To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 17 9:22:44 1999 Delivered-To: freebsd-smp@freebsd.org Received: from gndrsh.dnsmgr.net (GndRsh.dnsmgr.net [198.145.92.4]) by hub.freebsd.org (Postfix) with ESMTP id D0EBF1582B for ; Fri, 17 Sep 1999 09:22:31 -0700 (PDT) (envelope-from freebsd@gndrsh.dnsmgr.net) Received: (from freebsd@localhost) by gndrsh.dnsmgr.net (8.9.3/8.9.3) id JAA48657; Fri, 17 Sep 1999 09:21:33 -0700 (PDT) (envelope-from freebsd) From: "Rodney W. Grimes" Message-Id: <199909171621.JAA48657@gndrsh.dnsmgr.net> Subject: Re: Quad-PIII...exists? In-Reply-To: <37E2166A.35BE7A7C@surfusa.com> from Eb Farris at "Sep 17, 1999 06:22:34 am" To: efarris@surfusa.com (Eb Farris) Date: Fri, 17 Sep 1999 09:21:32 -0700 (PDT) Cc: dg@root.com, neil@synthcom.com (Neil Bradley), freebsd-smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL54 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I am surprised that none of you have tried Gigabyte. Been that route, did some of them. Infact let me see.. yep.. we are factory direct with them as well. Seems that they did some not so cool things in the past, lets see, 6 SIMM Triton chipset with screwy allowed combinations of single and double sided modules, caused us end user problems to no end (pun intended :-). Overall they make a good product, they pull stupid stunts like the above mentioned 6 SIMM thing to make it seem thier product is more expandable than others. > > We have been using Gigabyte for years and have had > great success with them. > > We ran into one BIOS problem that they fixed a couple > of years ago, but the quality and compatibility have > been great. Seems like I also had a major go around with them and Buslogic. -- Rod Grimes - KD7CAX - (RWG25) rgrimes@gndrsh.dnsmgr.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 17 10:38:46 1999 Delivered-To: freebsd-smp@freebsd.org Received: from pop3-3.enteract.com (pop3-3.enteract.com [207.229.143.32]) by hub.freebsd.org (Postfix) with SMTP id C249414E0A for ; Fri, 17 Sep 1999 10:38:29 -0700 (PDT) (envelope-from dscheidt@enteract.com) Received: (qmail 79247 invoked from network); 17 Sep 1999 17:38:26 -0000 Received: from shell-1.enteract.com (dscheidt@207.229.143.40) by pop3-3.enteract.com with SMTP; 17 Sep 1999 17:38:26 -0000 Date: Fri, 17 Sep 1999 12:38:26 -0500 (CDT) From: David Scheidt To: "Rodney W. Grimes" Cc: Eb Farris , dg@root.com, Neil Bradley , freebsd-smp@FreeBSD.ORG Subject: Re: Quad-PIII...exists? In-Reply-To: <199909171621.JAA48657@gndrsh.dnsmgr.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 17 Sep 1999, Rodney W. Grimes wrote: > > I am surprised that none of you have tried Gigabyte. I have a gigabyte 6BXD with two PII 400s in it. This is the one that hasn't got builtin SCSI or networking. It works very well; though, if I was going to do it again, I would likely get one of the ones with builtin SCSI. I don't know if it makes a decent server box, but as a workstation it rocks. I can do a massive CPU bound job in the background, and still get really zippy interactive performance. David Scheidt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 17 15:39:11 1999 Delivered-To: freebsd-smp@freebsd.org Received: from tahoe.cinenet.net (ns1.cinenet.net [198.147.76.65]) by hub.freebsd.org (Postfix) with ESMTP id 0550F14E91 for ; Fri, 17 Sep 1999 15:39:03 -0700 (PDT) (envelope-from sraja@cinenet.net) Received: from hermosa.cinenet.net (hermosa.cinenet.net [198.147.76.90]) by tahoe.cinenet.net (8.8.6/8.8.6) with SMTP id PAA22597 for ; Fri, 17 Sep 1999 15:39:02 -0700 (PDT) Date: Fri, 17 Sep 1999 15:39:01 -0700 (PDT) From: Suresh Rajagopalan To: freebsd-smp@FreeBSD.ORG Subject: SMP problems continue on 3.3-RC In-Reply-To: <199908301758.KAA16067@apollo.backplane.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Some weeks back I reported SMP problems with 3.2-STABLE. I'm continuing to have the same crash on 3.3-RC. This is on 3 seperate machines, NFS mounting from a Solaris NFS server. UP kernels don't seem to have this problem. -Suresh > :trace shows the following: > : > :panic (c0241d25,0,cc848cb0,1,c01769d4) at panic 0xa4 > :bsl1 (cc848c40, 2, cbbe6860, cbe3043f, cc848d00) at bs1 > :nfs_lookup (cbf3de30, cbafce00, cbf3dedc, cbf3deb80) at nfs_lookup 0x22f > :lookup(cbf3deb8,cbbe6860,cbbe6860,cbf3df94,cbf3de74) at lookup 0x2c1 > :namei(cbf3deb8,cbbe6860,c0292740,0,8162ce4) at namei 0x133 > :stat(cbbe6860,cbf3df94,13,5,bfbfdd50) at stat 0x44 > :syscall(27,bfbf0027,bfbfdd50,5,bfbfdb28) at syscall 0x107 > :Xint0x80_syscall() at Xint)x80_syscall + 0x4c To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 17 20:49:43 1999 Delivered-To: freebsd-smp@freebsd.org Received: from midten.fast.no (midten.fast.no [195.139.251.11]) by hub.freebsd.org (Postfix) with ESMTP id C8338154EC for ; Fri, 17 Sep 1999 20:49:12 -0700 (PDT) (envelope-from tegge@fast.no) Received: from fast.no (IDENT:tegge@midten.fast.no [195.139.251.11]) by midten.fast.no (8.9.3/8.9.3) with ESMTP id FAA19076; Sat, 18 Sep 1999 05:49:06 +0200 (CEST) Message-Id: <199909180349.FAA19076@midten.fast.no> To: sraja@cinenet.net Cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP problems continue on 3.3-RC From: Tor.Egge@fast.no In-Reply-To: Your message of "Fri, 17 Sep 1999 15:39:01 -0700 (PDT)" References: X-Mailer: Mew version 1.70 on Emacs 19.34.1 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Date: Sat, 18 Sep 1999 05:49:06 +0200 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > :panic (c0241d25,0,cc848cb0,1,c01769d4) at panic 0xa4 > > :bsl1 (cc848c40, 2, cbbe6860, cbe3043f, cc848d00) at bs1 > > :nfs_lookup (cbf3de30, cbafce00, cbf3dedc, cbf3deb80) at nfs_lookup 0x22f > > :lookup(cbf3deb8,cbbe6860,cbbe6860,cbf3df94,cbf3de74) at lookup 0x2c1 > > :namei(cbf3deb8,cbbe6860,c0292740,0,8162ce4) at namei 0x133 > > :stat(cbbe6860,cbf3df94,13,5,bfbfdd50) at stat 0x44 > > :syscall(27,bfbf0027,bfbfdd50,5,bfbfdb28) at syscall 0x107 > > :Xint0x80_syscall() at Xint)x80_syscall + 0x4c nfs_lookup called vget, but the trace is incomplete due to s_lock being a frameless function. One of the simple_lock calls in vget failed due to the vnode interlock already being held by the same CPU. If it was held by any other CPU then you would get a hang instead of a panic. I suggest testing a UP kernel with the SIMPLELOCK_DEBUG kernel option defined and mi_switch modified to panic instead of just printing a warning when a process attempts to sleep with a simple lock held. - Tor Egge To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 17 23: 5:40 1999 Delivered-To: freebsd-smp@freebsd.org Received: from tahoe.cinenet.net (ns1.cinenet.net [198.147.76.65]) by hub.freebsd.org (Postfix) with ESMTP id C834215A14 for ; Fri, 17 Sep 1999 23:05:29 -0700 (PDT) (envelope-from sraja@cinenet.net) Received: from hermosa.cinenet.net (hermosa.cinenet.net [198.147.76.90]) by tahoe.cinenet.net (8.8.6/8.8.6) with SMTP id XAA04422; Fri, 17 Sep 1999 23:05:24 -0700 (PDT) Date: Fri, 17 Sep 1999 23:05:24 -0700 (PDT) From: Suresh Rajagopalan To: Tor.Egge@fast.no Cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP problems continue on 3.3-RC In-Reply-To: <199909180349.FAA19076@midten.fast.no> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I suggest testing a UP kernel with the SIMPLELOCK_DEBUG kernel option > defined and mi_switch modified to panic instead of just printing a > warning when a process attempts to sleep with a simple lock held. > Most of the time I get a hang, not a panic. I'll try your suggestion. Where is the mi_switch defined? Thanks Suresh To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 18 7:51:36 1999 Delivered-To: freebsd-smp@freebsd.org Received: from atlrel1.hp.com (atlrel1.hp.com [156.153.255.210]) by hub.freebsd.org (Postfix) with ESMTP id C952A14D43 for ; Sat, 18 Sep 1999 07:51:10 -0700 (PDT) (envelope-from darrylo@sr.hp.com) Received: from postal.sr.hp.com (root@postal.sr.hp.com [15.4.46.173]) by atlrel1.hp.com (8.8.6 (PHNE_17135)/8.8.5tis) with ESMTP id KAA16742; Sat, 18 Sep 1999 10:49:53 -0400 (EDT) Received: from mina.sr.hp.com (root@mina.sr.hp.com [15.4.42.247]) by postal.sr.hp.com with ESMTP (8.8.6 (PHNE_17190)/8.7.3 TIS 5.0) id HAA12026; Sat, 18 Sep 1999 07:50:08 -0700 (PDT) Received: from localhost (darrylo@mina.sr.hp.com [15.4.42.247]) by mina.sr.hp.com with ESMTP (8.8.6 (PHNE_17135)/8.7.3 TIS 5.0) id HAA25704; Sat, 18 Sep 1999 07:50:23 -0700 (PDT) Message-Id: <199909181450.HAA25704@mina.sr.hp.com> To: David Scheidt Cc: "Rodney W. Grimes" , Eb Farris , dg@root.com, Neil Bradley , freebsd-smp@FreeBSD.ORG Subject: Re: Quad-PIII...exists? Reply-To: Darryl Okahata In-reply-to: Your message of "Fri, 17 Sep 1999 12:38:26 CDT." Mime-Version: 1.0 (generated by tm-edit 7.108) Content-Type: text/plain; charset=US-ASCII Date: Sat, 18 Sep 1999 07:50:23 -0700 From: Darryl Okahata Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org David Scheidt wrote: > I have a gigabyte 6BXD with two PII 400s in it. This is the one that > hasn't got builtin SCSI or networking. It works very well; though, if I was > going to do it again, I would likely get one of the ones with builtin SCSI. > I don't know if it makes a decent server box, but as a workstation it > rocks. I can do a massive CPU bound job in the background, and still get > really zippy interactive performance. Here's another plug for Gigabyte. I've got the 6BXDS (6BXD w/7895 SCSI), and it's very nice. If you don't need LVD or U2 and can live with narrow or plain wide SCSI (both busses can be used simultaneously), it's a pretty good deal (around US$300, today, although it used to be around US$230 a few months back). The 6BXD is also one of the cheapest dual-processor BX-based boards, being around US$160. If you don't need SCSI, I think it's a great way to go. It is difficult to find, though. -- Darryl Okahata darrylo@sr.hp.com DISCLAIMER: this message is the author's personal opinion and does not constitute the support, opinion, or policy of Hewlett-Packard, or of the little green men that have been following him all day. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message