From owner-freebsd-smp Sun Oct 6 11:58:35 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8619837B401; Sun, 6 Oct 2002 11:58:33 -0700 (PDT) Received: from out017.verizon.net (out017pub.verizon.net [206.46.170.94]) by mx1.FreeBSD.org (Postfix) with ESMTP id B646043E4A; Sun, 6 Oct 2002 11:58:32 -0700 (PDT) (envelope-from res03db2@verizon.net) Received: from verizon.net ([4.47.70.146]) by out017.verizon.net (InterMail vM.5.01.05.09 201-253-122-126-109-20020611) with ESMTP id <20021006185831.CUFU6394.out017.verizon.net@verizon.net>; Sun, 6 Oct 2002 13:58:31 -0500 Received: (from res03db2@localhost) by verizon.net (8.9.3/8.9.3) id LAA29059; Sun, 6 Oct 2002 11:58:16 -0700 (PDT) (envelope-from res03db2) Date: Sun, 6 Oct 2002 11:58:16 -0700 From: Robert Clark To: Terry Lambert Cc: Nate Lawson , David Francheski , freebsd-arch@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: Running independent kernel instances on dual-Xeon/E7500 system Message-ID: <20021006115816.A28963@darkstar.gte.net> References: <3D9EB0A4.4CD09E20@mindspring.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.4i In-Reply-To: <3D9EB0A4.4CD09E20@mindspring.com>; from tlambert2@mindspring.com on Sat, Oct 05, 2002 at 02:28:04AM -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I've often thought it would be nice to be able to devote one processor to a RT style OS instance that continuous duty doing "throw away" work updating the display, audio, etc. Using a general purpose CPU for graphics and sound work may not result in the kinds of performance you get with a GPU, but I have to imagine it would have a better chance of encouraging "free" driver development. On the flip side, the OS instance that didn't have anything to do with audio/video could spend more of its time doing network/disk I/O, and more traditional duties. [RC] On Sat, Oct 05, 2002 at 02:28:04AM -0700, Terry Lambert wrote: > Nate Lawson wrote: > > On Fri, 4 Oct 2002, David Francheski wrote: > > > I have a dual-Xeon processor (with E7500 chipset) motherboard. > > > Can anybody tell me what the development effort would be to > > > boot and run two independent copies of the FreeBSD kernel, > > > one on each Xeon processor? By this I mean that an SMP > > > enabled kernel would not be utilized, each kernel would be UP. > > > > > > Regards, > > > David L. Francheski > > > > Not possible without another BIOS, PCI bus, and separate memory -- > > i.e. another PC. > > IPL'ing is not the same as "running". So long as you crafted the > memory image of the second OS and its page tables, etc., using the > first processor, there should be no problem running a second copy > of an OS on an AP, as a result of a START IPI from the BP, after > the code is crafted. Thus there is no need for a separate BIOS. > > -- > > I've personally considered pursuing the ability to run code seperately, > though with the same 4G address space, seperated, so as to permit > running a debugger against a "crashed" FreeBSD "system" running on an > AP, doing the debugging from the BP, as a hosted system. The cost > in labor would be 2-3 months of continuous work, I think... that is > the estimate I arrived at, when I considered the project previously. > Doing this certaily beats the cost of buying an ICE to get similar > capability. > > > It would be interesting to see what other people have to say on this, > other than "can't be done" (not to pick on you in particular, here; > this is the knee-jerk reaction many people have to things like this). > > -- Terry > > 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 Sun Oct 6 17:16:30 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C434837B401; Sun, 6 Oct 2002 17:16:29 -0700 (PDT) Received: from flamingo.mail.pas.earthlink.net (flamingo.mail.pas.earthlink.net [207.217.120.232]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5F41743E97; Sun, 6 Oct 2002 17:16:29 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from pool0162.cvx22-bradley.dialup.earthlink.net ([209.179.198.162] helo=mindspring.com) by flamingo.mail.pas.earthlink.net with esmtp (Exim 3.33 #1) id 17yLZS-0004wa-00; Sun, 06 Oct 2002 17:16:26 -0700 Message-ID: <3DA0D20D.C47E4EF8@mindspring.com> Date: Sun, 06 Oct 2002 17:15:09 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Antony T Curtis Cc: Nate Lawson , David Francheski , freebsd-arch@FreeBSD.ORG, freebsd-smp@FreeBSD.org Subject: Re: Running independent kernel instances on dual-Xeon/E7500 system References: <3D9EB0A4.4CD09E20@mindspring.com> <3D9EF6E9.9040700@ntlworld.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Nathan Hawkins wrote: > Each approach has advantages and disadvantages, depending on what you're > trying to accomplish. I think in this case, it's that benchmarked performance actually goes down in FreeBSD 4.6 when you run SMP, as opposed to running UP, and FreeBSD -current is even worse, even if you disable the debugging that's on by default. Tools like "netperf" aren't really capable of taking advantage of additional processors, but they are excellent at showing the incremental slowdown that results from lack of CPU affinity (if applicable), as well as any additional locking overhead (if applicable). Tools that run against web servers, where the web server has been written to run with multiple processes (or mutithreaded, if the threads system on the platform is SMP scalable) show less improvement than expected; e.g.: http://www.softwareqatest.com/qatweb1.html#LOAD ...but they will at least show some small improvement with SMP. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Oct 6 19:48:46 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9279737B401 for ; Sun, 6 Oct 2002 19:48:43 -0700 (PDT) Received: from rootlabs.com (root.org [67.118.192.226]) by mx1.FreeBSD.org (Postfix) with SMTP id 0D79343E7B for ; Sun, 6 Oct 2002 19:48:39 -0700 (PDT) (envelope-from nate@rootlabs.com) Received: (qmail 5551 invoked by uid 1000); 7 Oct 2002 02:48:40 -0000 Date: Sun, 6 Oct 2002 19:48:40 -0700 (PDT) From: Nate Lawson To: freebsd-arch@FreeBSD.ORG Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Running independent kernel instances on dual-Xeon/E7500 system In-Reply-To: <20021006115816.A28963@darkstar.gte.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Sorry for the unhelpful first posting. I sent a more detailed letter via private mail, recommending he look into the exokernel papers. My dismissiveness was due to anticipating the direction this was going, which is nicely shown by the response below. In short, dedicated processors for IO were used in the minicomputer days but are wasteful nowadays when you have lightweight interrupts and/or polling when appropriate. If your scheduler sucks, fix it. If a device needs extra processing equivalent to another N Ghz CPU, the vendor will add silicon. The "S" in SMP is for symmetric, lest we forget. -Nate On Sun, 6 Oct 2002, Robert Clark wrote: > I've often thought it would be nice to be able to devote > one processor to a RT style OS instance that continuous > duty doing "throw away" work updating the display, audio, > etc. > > Using a general purpose CPU for graphics and sound work > may not result in the kinds of performance you get with > a GPU, but I have to imagine it would have a better > chance of encouraging "free" driver development. > > On the flip side, the OS instance that didn't have > anything to do with audio/video could spend more of > its time doing network/disk I/O, and more traditional > duties. > > [RC] > > On Sat, Oct 05, 2002 at 02:28:04AM -0700, Terry Lambert wrote: > > Nate Lawson wrote: > > > On Fri, 4 Oct 2002, David Francheski wrote: > > > > I have a dual-Xeon processor (with E7500 chipset) motherboard. > > > > Can anybody tell me what the development effort would be to > > > > boot and run two independent copies of the FreeBSD kernel, > > > > one on each Xeon processor? By this I mean that an SMP > > > > enabled kernel would not be utilized, each kernel would be UP. > > > > > > > > Regards, > > > > David L. Francheski > > > > > > Not possible without another BIOS, PCI bus, and separate memory -- > > > i.e. another PC. > > > > IPL'ing is not the same as "running". So long as you crafted the > > memory image of the second OS and its page tables, etc., using the > > first processor, there should be no problem running a second copy > > of an OS on an AP, as a result of a START IPI from the BP, after > > the code is crafted. Thus there is no need for a separate BIOS. > > > > > > > -- > > > > I've personally considered pursuing the ability to run code seperately, > > though with the same 4G address space, seperated, so as to permit > > running a debugger against a "crashed" FreeBSD "system" running on an > > AP, doing the debugging from the BP, as a hosted system. The cost > > in labor would be 2-3 months of continuous work, I think... that is > > the estimate I arrived at, when I considered the project previously. > > Doing this certaily beats the cost of buying an ICE to get similar > > capability. > > > > > > It would be interesting to see what other people have to say on this, > > other than "can't be done" (not to pick on you in particular, here; > > this is the knee-jerk reaction many people have to things like this). > > > > -- Terry > > > > 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-arch" 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 Sun Oct 6 21:12: 5 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D151037B408; Sun, 6 Oct 2002 21:12:03 -0700 (PDT) Received: from gull.mail.pas.earthlink.net (gull.mail.pas.earthlink.net [207.217.120.84]) by mx1.FreeBSD.org (Postfix) with ESMTP id 70BA043E81; Sun, 6 Oct 2002 21:12:03 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from pool0276.cvx22-bradley.dialup.earthlink.net ([209.179.199.21] helo=mindspring.com) by gull.mail.pas.earthlink.net with esmtp (Exim 3.33 #1) id 17yPFR-0007ci-00; Sun, 06 Oct 2002 21:12:02 -0700 Message-ID: <3DA10949.218488B9@mindspring.com> Date: Sun, 06 Oct 2002 21:10:49 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Nate Lawson Cc: freebsd-arch@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: Running independent kernel instances on dual-Xeon/E7500 system References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Nate Lawson wrote: > My dismissiveness was due to anticipating the direction this was going, > which is nicely shown by the response below. In short, dedicated > processors for IO were used in the minicomputer days but are wasteful > nowadays when you have lightweight interrupts and/or polling when > appropriate. Yet, I keep running into employers who want to pay people to do exactly that, particularly for offloading network processing to one processor, and running applications on the other. And then there's the Tigon II firmware rewrite for FreeBSD, to offload interrupt and copy processing. And CGD's work for Sibytes (NetBSD 64bit MIPS-based network coprocessor board) doing just that got the company sold to Broadcom for what, $700M? 8-). > If your scheduler sucks, fix it. If a device needs extra processing > equivalent to another N Ghz CPU, the vendor will add silicon. The "S" in > SMP is for symmetric, lest we forget. People keep saying that, and then keep not running interrupts in virtual wire mode, so that their delivery is "S" as in "symmetric"... ;^). Actually, NT proved that wiring particular interrupts to particular processors was the way to go -- that was one of the things they did to beat the Linux numbers in both the Netcraft and Ziff-Davis benchmarks... perfect symmetry isn't all that it's promised. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Oct 6 21:56:30 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8BD4937B404 for ; Sun, 6 Oct 2002 21:56:28 -0700 (PDT) Received: from rootlabs.com (root.org [67.118.192.226]) by mx1.FreeBSD.org (Postfix) with SMTP id 2473F43E65 for ; Sun, 6 Oct 2002 21:56:24 -0700 (PDT) (envelope-from nate@rootlabs.com) Received: (qmail 5885 invoked by uid 1000); 7 Oct 2002 04:56:26 -0000 Date: Sun, 6 Oct 2002 21:56:26 -0700 (PDT) From: Nate Lawson To: freebsd-arch@FreeBSD.ORG Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Running independent kernel instances on dual-Xeon/E7500 system In-Reply-To: <3DA10949.218488B9@mindspring.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Sun, 6 Oct 2002, Terry Lambert wrote: > Nate Lawson wrote: > > My dismissiveness was due to anticipating the direction this was going, > > which is nicely shown by the response below. In short, dedicated > > processors for IO were used in the minicomputer days but are wasteful > > nowadays when you have lightweight interrupts and/or polling when > > appropriate. > > Yet, I keep running into employers who want to pay people to do > exactly that, particularly for offloading network processing to > one processor, and running applications on the other. Been there, hence the touchiness. > And then there's the Tigon II firmware rewrite for FreeBSD, to > offload interrupt and copy processing. And CGD's work for Sibytes > (NetBSD 64bit MIPS-based network coprocessor board) doing just that > got the company sold to Broadcom for what, $700M? > > 8-). I agree that when there are spare cycles available on the _device_'s processor it should be doing more work. But that's very different from dedicating a processor that could otherwise be doing useful work (given a well-written SMP-aware OS of course). > > If your scheduler sucks, fix it. If a device needs extra processing > > equivalent to another N Ghz CPU, the vendor will add silicon. The "S" in > > SMP is for symmetric, lest we forget. > > People keep saying that, and then keep not running interrupts in > virtual wire mode, so that their delivery is "S" as in "symmetric"... > ;^). > > Actually, NT proved that wiring particular interrupts to particular > processors was the way to go -- that was one of the things they did > to beat the Linux numbers in both the Netcraft and Ziff-Davis > benchmarks... perfect symmetry isn't all that it's promised. > > -- Terry I'm not sure that breaks my definition of symmetric since that sounds like they were just setting the processor affinity per interrupt. ;-) I agree that for a given fixed workload profile, it may make sense to build a single-purpose device out of off-the-shelf parts. But most of the time, the decision to go that way is the result of a knee-jerk reaction that "I'm a Real Programmer and I want bare metal because it's faster". I believe that mostly results in slower systems because the workload always changes out from under the designer's assumptions. Hence we get more real benefits from versatile things like branch prediction, register scoreboarding, etc. that make their decisions at runtime instead of chalkboard time. Here's an archived email by VJ that seems amazingly relevant, even today: http://www.root.org/ip-development/news/vanj.88jul20.txt -Nate To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Oct 7 19: 6:31 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 293B737B401; Mon, 7 Oct 2002 19:06:28 -0700 (PDT) Received: from out011.verizon.net (out011pub.verizon.net [206.46.170.135]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7655C43E65; Mon, 7 Oct 2002 19:06:27 -0700 (PDT) (envelope-from res03db2@verizon.net) Received: from verizon.net ([4.47.70.146]) by out011.verizon.net (InterMail vM.5.01.05.09 201-253-122-126-109-20020611) with ESMTP id <20021008020626.IVCC17563.out011.verizon.net@verizon.net>; Mon, 7 Oct 2002 21:06:26 -0500 Received: (from res03db2@localhost) by verizon.net (8.9.3/8.9.3) id TAA31352; Mon, 7 Oct 2002 19:06:15 -0700 (PDT) (envelope-from res03db2) Date: Mon, 7 Oct 2002 19:06:10 -0700 From: Robert Clark To: Nate Lawson Cc: freebsd-arch@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: Running independent kernel instances on dual-Xeon/E7500 system Message-ID: <20021007190610.A31292@darkstar.gte.net> References: <20021006115816.A28963@darkstar.gte.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.4i In-Reply-To: ; from nate@root.org on Sun, Oct 06, 2002 at 07:48:40PM -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Ok, if I missed the clue boat, where does it call next? I'd like to catch it before it gets too far away. Is FreeBSD getting better on SMP? [RC] On Sun, Oct 06, 2002 at 07:48:40PM -0700, Nate Lawson wrote: > Sorry for the unhelpful first posting. I sent a more detailed letter via > private mail, recommending he look into the exokernel papers. > > My dismissiveness was due to anticipating the direction this was going, > which is nicely shown by the response below. In short, dedicated > processors for IO were used in the minicomputer days but are wasteful > nowadays when you have lightweight interrupts and/or polling when > appropriate. > > If your scheduler sucks, fix it. If a device needs extra processing > equivalent to another N Ghz CPU, the vendor will add silicon. The "S" in > SMP is for symmetric, lest we forget. > > -Nate > > On Sun, 6 Oct 2002, Robert Clark wrote: > > I've often thought it would be nice to be able to devote > > one processor to a RT style OS instance that continuous > > duty doing "throw away" work updating the display, audio, > > etc. > > > > Using a general purpose CPU for graphics and sound work > > may not result in the kinds of performance you get with > > a GPU, but I have to imagine it would have a better > > chance of encouraging "free" driver development. > > > > On the flip side, the OS instance that didn't have > > anything to do with audio/video could spend more of > > its time doing network/disk I/O, and more traditional > > duties. > > > > [RC] > > > > On Sat, Oct 05, 2002 at 02:28:04AM -0700, Terry Lambert wrote: > > > Nate Lawson wrote: > > > > On Fri, 4 Oct 2002, David Francheski wrote: > > > > > I have a dual-Xeon processor (with E7500 chipset) motherboard. > > > > > Can anybody tell me what the development effort would be to > > > > > boot and run two independent copies of the FreeBSD kernel, > > > > > one on each Xeon processor? By this I mean that an SMP > > > > > enabled kernel would not be utilized, each kernel would be UP. > > > > > > > > > > Regards, > > > > > David L. Francheski > > > > > > > > Not possible without another BIOS, PCI bus, and separate memory -- > > > > i.e. another PC. > > > > > > IPL'ing is not the same as "running". So long as you crafted the > > > memory image of the second OS and its page tables, etc., using the > > > first processor, there should be no problem running a second copy > > > of an OS on an AP, as a result of a START IPI from the BP, after > > > the code is crafted. Thus there is no need for a separate BIOS. > > > > > > > > > > > > -- > > > > > > I've personally considered pursuing the ability to run code seperately, > > > though with the same 4G address space, seperated, so as to permit > > > running a debugger against a "crashed" FreeBSD "system" running on an > > > AP, doing the debugging from the BP, as a hosted system. The cost > > > in labor would be 2-3 months of continuous work, I think... that is > > > the estimate I arrived at, when I considered the project previously. > > > Doing this certaily beats the cost of buying an ICE to get similar > > > capability. > > > > > > > > > It would be interesting to see what other people have to say on this, > > > other than "can't be done" (not to pick on you in particular, here; > > > this is the knee-jerk reaction many people have to things like this). > > > > > > -- Terry > > > > > > 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-arch" in the body of the message > > > > > 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 Mon Oct 7 20:41:30 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id F397137B401 for ; Mon, 7 Oct 2002 20:41:27 -0700 (PDT) Received: from mailhub.cns.ksu.edu (grunt.ksu.ksu.edu [129.130.12.17]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1ADED43E4A for ; Mon, 7 Oct 2002 20:41:27 -0700 (PDT) (envelope-from jrf4772@ksu.edu) Received: from unix1.cc.ksu.edu (jrf4772@unix1.cc.ksu.edu [129.130.12.3]) by mailhub.cns.ksu.edu (8.9.1/8.9.1/mailhub+tar) with ESMTP id WAA06985; Mon, 7 Oct 2002 22:41:07 -0500 (CDT) Received: from localhost (jrf4772@localhost) by unix1.cc.ksu.edu (8.8.8+Sun/8.8.8) with ESMTP id WAA06259; Mon, 7 Oct 2002 22:41:06 -0500 (CDT) X-Authentication-Warning: unix1.cc.ksu.edu: jrf4772 owned process doing -bs Date: Mon, 7 Oct 2002 22:41:05 -0500 (CDT) From: Jonathan R Feldkamp X-X-Sender: To: Robert Bopko Cc: Danny Braniss , Stephen Karrington , James Schmidt , Subject: Re: se7500cw2 hack to try In-Reply-To: <20021006100955.GA13526@finom.estimese.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Danny, I tried your new version of mp_machdep.c, it didn't compile... :) lines 2205, and 2206 use an undefined variable 'x', I replaced them with logical_cpu, compiled, installed, and booted. the results are below, smp is still dead :( ---- dmesg stuff ------------------------ Programming 24 pins in IOAPIC #0 IOAPIC #0 intpin 2 -> irq 0 Programming 24 pins in IOAPIC #1 Programming 24 pins in IOAPIC #2 CPU0: cpu_apic_version[0] is 0x00000000 CPU1: cpu_apic_version[0] is 0x00050014 FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 0, version: 0x00050014, at 0xfee00000 cpu1 (AP): apic id: 3, version: 0x00050014, at 0xfee00000 io0 (APIC): apic id: 2, version: 0x00178020, at 0xfec00000 io1 (APIC): apic id: 3, version: 0x00178020, at 0xfec80000 io2 (APIC): apic id: 4, version: 0x00178020, at 0xfec80400 Preloaded elf kernel "kernel" at 0xc03b2000. ------------------------------------------ ---- sysctl stuff ------------------------ osage:~>sysctl machdep.smp_active machdep.smp_active: 0 osage:~>sysctl machdep.smp_cpus machdep.smp_cpus: 1 osage:~>sysctl hw.ncpu hw.ncpu: 7531633 ------------------------------------------ for those at smp@freebsd.org, I put a copy Danny's mp_machdep.c up at: http://www-personal.ksu.edu/~jrf4772/mp_machdep.c I think we may be getting close here... Thanks, jon On Sun, 6 Oct 2002, Robert Bopko wrote: > ahh i am not a smp specialist. :-( > > dmesg from working linux: > > Initializing CPU#1 > masked ExtINT on CPU#1 > ESR value before enabling vector: 00000000 > ESR value after enabling vector: 00000000 > Calibrating delay loop... 3984.58 BogoMIPS > CPU: Before vendor init, caps: 3febfbff 00000000 00000000, vendor = 0 > CPU: L1 I cache: 12K, L1 D cache: 8K > CPU: L2 cache: 512K > !!!!CPU: Physical Processor ID: 3!!!! > CPU: After vendor init, caps: 3febfbff 00000000 00000000 00000000 > Intel machine check reporting enabled on CPU#1. > CPU: After generic, caps: 3febfbff 00000000 00000000 00000000 > CPU: Common caps: 3febfbff 00000000 00000000 00000000 > CPU1: Intel(R) XEON(TM) CPU 2.00GHz stepping 04 > Total of 2 processors activated (7956.07 BogoMIPS). > > but the panic message: > > "AP #1 (PHY# 6) failed!" > so the freebsd kernel want use Physical ID 6 not ID 3. > > here is another mp_machdep.c with overrided AP PHY ID to 3, and > disabled second IPI startup instructions. > and please add another kernel parameter to loader.conf: > hw.ncpu="2" > > thanks. > > ps: hw.ncpu: 7531970? :-) your pc must be big like las vegas. ;-) > > On Sun, Oct 06, 2002 at 12:27:12PM +0300, Danny Braniss wrote: > > > > > now please. > > > edit /boot/loader.conf and add there this lines: > > > machdep.smp_active="1" > > > machdep.smp_cpus="2" > > > > no luck, > > hw.ncpu: 7531970 > > machdep.smp_cpus: 1 > > machdep.smp_active: 0 > > > > as long as we are debugging, this is ok, putting things in loader.conf, > > but in my case most of these boxes are 'diskless/dataless' with a shared > > root, and not all are smp. > > > > danny > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Oct 7 20:59:14 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 127B337B401 for ; Mon, 7 Oct 2002 20:59:14 -0700 (PDT) Received: from mailhub.cns.ksu.edu (grunt.ksu.ksu.edu [129.130.12.17]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7A98443E86 for ; Mon, 7 Oct 2002 20:59:13 -0700 (PDT) (envelope-from jrf4772@ksu.edu) Received: from klweb04 (direct-webmail [10.0.12.2]) by mailhub.cns.ksu.edu (8.9.1/8.9.1/mailhub+tar) with ESMTP id WAA15076; Mon, 7 Oct 2002 22:59:12 -0500 (CDT) X-WebMail-UserID: jrf4772 Date: Mon, 7 Oct 2002 22:59:12 -0500 From: jrf4772 To: smp@freebsd.org Cc: Danny Braniss , James Schmidt , Robert Bopko , Stephen Karrington X-EXP32-SerialNo: 00002882 Subject: RE: se7500cw2 hack to try Message-ID: <3DA7378A@klweb04> Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit X-Mailer: WebMail (Hydra) SMTP v3.61 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I apologize, it was Robert's code, not Danny's Although that hack didn't fix smp, it did seem to make X a little unstable. thanks again, jon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Oct 7 22:24:54 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5705837B401 for ; Mon, 7 Oct 2002 22:24:53 -0700 (PDT) Received: from cse.cs.huji.ac.il (cse.cs.huji.ac.il [132.65.16.30]) by mx1.FreeBSD.org (Postfix) with ESMTP id CE8E543E81 for ; Mon, 7 Oct 2002 22:24:52 -0700 (PDT) (envelope-from danny@cs.huji.ac.il) Received: from pampa.cs.huji.ac.il ([132.65.80.32] ident=danny) by cse.cs.huji.ac.il with esmtp id 17ymqA-000Pmr-00; Tue, 08 Oct 2002 07:23:30 +0200 X-Mailer: exmh version 2.5 07/13/2001 with nmh-1.0.4 To: Jonathan R Feldkamp Cc: Robert Bopko , Stephen Karrington , James Schmidt , smp@freebsd.org Subject: Re: se7500cw2 hack to try In-Reply-To: Message from Jonathan R Feldkamp of "Mon, 07 Oct 2002 22:41:05 EST." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 08 Oct 2002 07:23:30 +0200 From: Danny Braniss Message-Id: Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org well, the bright side is that it's not panicking :-) and yes, the code belongs to Robert. danny To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Oct 9 17:49:48 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1997C37B401 for ; Wed, 9 Oct 2002 17:49:46 -0700 (PDT) Received: from ultimate.com (h525405f6161e.ne.client2.attbi.com [66.30.204.193]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7CD3143E42 for ; Wed, 9 Oct 2002 17:49:45 -0700 (PDT) (envelope-from phil@ultimate.com) Received: (from phil@localhost) by ultimate.com (8.12.0.Beta7/8.12.0.Beta7) id g9A0nhDF012890; Wed, 9 Oct 2002 20:49:43 -0400 (EDT) Date: Wed, 9 Oct 2002 20:49:43 -0400 (EDT) From: Phil Budne Message-Id: <200210100049.g9A0nhDF012890@ultimate.com> To: freebsd-smp@freebsd.org Subject: Re: -STABLE painfully slow on multi-processor (followup) Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Last week I posted about an SMP system with an intel "Lancewood" (L440GX) motherboard that had putrid performance when booted with uniprocessor 4.5, but was slightly better when booted under an SMP kernel (4.5 or 4.7RC). I've since found that the problem is that the boot processor is slow (100x slower than it ought to be), while the second processor seems fine. I was able to buildworld in a reasonable amount of time if I keep the primary processor busy (I wrote a program that runs at realtime priority which spins, and if it took too long (on slow processor), keeps spinning, otherwise it gives up the (good) processor. sched_yield() wasn't doing the trick, so I used usleep(1)). I also cracked open the box that booted 4.5 (and 4.7) and only reported one processor -- it has two (also shown in BIOS POST). A similar problem was previously reported here (second processor disappeared after upgrading from 4.1 to 4.5); http://docs.freebsd.org/cgi/getmsg.cgi?fetch=90892+0+archive/2002/freebsd-smp/20020120.freebsd-smp The boxes were put together by VALinux, and while I'm not the paranoid sort (not while I'm on my medication anyhow), I am willing to believe it's a BIOS problem that isn't exposed by Linux or FreeBSD 4.1. There are two more (presumably identical) boxes still running Linux (2.2.16 and 2.2.19), which seem fine. Please cc: me on any replies! Thanks! Phil Budne Consultant To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Oct 10 0:18:28 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1F5D637B401 for ; Thu, 10 Oct 2002 00:18:27 -0700 (PDT) Received: from woolridge.ca (H10.C245.tor.velocet.net [216.138.245.10]) by mx1.FreeBSD.org (Postfix) with SMTP id ECA5243E91 for ; Thu, 10 Oct 2002 00:18:25 -0700 (PDT) (envelope-from dale-list-freebsd-smp@woolridge.org) Received: (qmail 78983 invoked by uid 1000); 10 Oct 2002 07:18:25 -0000 Date: Thu, 10 Oct 2002 03:18:25 -0400 From: Dale Woolridge To: freebsd-smp@FreeBSD.ORG Cc: Yoriaki FUJIMORI Subject: Re: dnetc on xeon Message-ID: <20021010071825.GH539@woolridge.ca> Mail-Followup-To: freebsd-smp@FreeBSD.ORG, Yoriaki FUJIMORI References: <200210031237.g93CbmQR014892@grafin.fujimori.cache.waseda.ac.jp> <3375.1033650434@critter.freebsd.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3375.1033650434@critter.freebsd.dk> User-Agent: Mutt/1.5.1i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On 3-Oct-2002 15:07 Poul-Henning Kamp wrote: | In message <200210031237.g93CbmQR014892@grafin.fujimori.cache.waseda.ac.jp>, Yo | riaki FUJIMORI writes: | | >My question is if anybody of you have experienced similar symptom | >on their xeon dual boxes. | >In advance, thank you for your attention. | | Your hardware is not up to snuff. | | In order of probablity: | | 1. You were not by any chance overclocking ? | | 2. Is your powersupply sufficient ? | | 3. Is your cooling sufficient ? I've been experiencing similar problems with my dual p3 (on a vp6) system. I couldn't build a kernel, the world, and some ports (large ones usually). I happened to be running one of the cpuburn port programs (burnBX) to see if it would crap out; meanwhile I was building a kernel. I was able to build the kernel without these strange compile errors. Also, burnBX would die (exit status -1), but certainly not predictably. In my case, I'm not overclocking (and never have with this board), have no powersupply problems, and have sufficient cooling (my cpus run at about 30-33 celsius). A friend suggested disabling my L2 cache, which I did and all my compilation problems have gone away. Of course, I still can't get SMP up and running, but that's a whole other story (maybe another hardware issue). At this point, I'm not sure if it's the processor itself (unlikely I'd guess) or a mobo problem (more likely), so if anyone has suggestions I'd really appreciate them. -- -Dale To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Oct 10 2: 3:12 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E625537B401 for ; Thu, 10 Oct 2002 02:03:10 -0700 (PDT) Received: from anchor-post-39.mail.demon.net (anchor-post-39.mail.demon.net [194.217.242.80]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7CB4C43E9C for ; Thu, 10 Oct 2002 02:03:09 -0700 (PDT) (envelope-from michaelossareh@onetel.net.uk) Received: from dayjah.demon.co.uk ([62.49.68.53] helo=maggie.mossareh.home.org) by anchor-post-39.mail.demon.net with smtp (Exim 3.36 #2) id 17zZDo-0001cM-0U for freebsd-smp@FreeBSD.ORG; Thu, 10 Oct 2002 10:03:08 +0100 Date: Thu, 10 Oct 2002 10:08:23 +0000 From: Michael Ossareh To: freebsd-smp@FreeBSD.ORG Subject: Abit SMP boards under FreeBSD (was Re: dnetc on xeon) Message-Id: <20021010100823.0a1e00ad.michaelossareh@onetel.net.uk> In-Reply-To: <20021010071825.GH539@woolridge.ca> References: <200210031237.g93CbmQR014892@grafin.fujimori.cache.waseda.ac.jp> <3375.1033650434@critter.freebsd.dk> <20021010071825.GH539@woolridge.ca> X-Mailer: Sylpheed version 0.8.2 (GTK+ 1.2.10; i686-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Thu, 10 Oct 2002 03:18:25 -0400 Dale Woolridge wrote: > On 3-Oct-2002 15:07 Poul-Henning Kamp wrote: > | In message <200210031237.g93CbmQR014892@grafin.fujimori.cache.waseda.ac.jp>, Yo > | riaki FUJIMORI writes: > | > | >My question is if anybody of you have experienced similar symptom > | >on their xeon dual boxes. > | >In advance, thank you for your attention. > | > | Your hardware is not up to snuff. > | > | In order of probablity: > | > | 1. You were not by any chance overclocking ? > | > | 2. Is your powersupply sufficient ? > | > | 3. Is your cooling sufficient ? > > > I've been experiencing similar problems with my dual p3 (on a vp6) system. > I couldn't build a kernel, the world, and some ports (large ones usually). > I happened to be running one of the cpuburn port programs (burnBX) to see > if it would crap out; meanwhile I was building a kernel. I was able to > build the kernel without these strange compile errors. Also, burnBX would > die (exit status -1), but certainly not predictably. > > In my case, I'm not overclocking (and never have with this board), have > no powersupply problems, and have sufficient cooling (my cpus run at > about 30-33 celsius). > > A friend suggested disabling my L2 cache, which I did and all my compilation > problems have gone away. Of course, I still can't get SMP up and running, > but that's a whole other story (maybe another hardware issue). At this > point, I'm not sure if it's the processor itself (unlikely I'd guess) or > a mobo problem (more likely), so if anyone has suggestions I'd really > appreciate them. > -- > -Dale > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message My Abit BP6 had similar problems, but only whilst I was overclocking. Thanks to the great guys in #freebsd for advising me to unoverclock the compile problems went away. I believe the Abit VP6 and BP6 are quite similar and the chipset runs very hot when in SMP mode, so since its not a proper server board you might want to invest in some active cooling for your chipset (usually just has a green heat sink on it). Michael Ossareh To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Oct 10 14:38:46 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4F9DA37B401 for ; Thu, 10 Oct 2002 14:38:45 -0700 (PDT) Received: from smtp2.electric.net (smtp2.electric.net [216.129.90.15]) by mx1.FreeBSD.org (Postfix) with ESMTP id DC81B43E9C for ; Thu, 10 Oct 2002 14:38:44 -0700 (PDT) (envelope-from davidf@caymas.com) Received: from zhora.electric.net ([216.129.90.89]) by smtp2.electric.net with smtp (Exim 4.04) id 17zl0w-00044P-01 for freebsd-smp@freebsd.org; Thu, 10 Oct 2002 14:38:38 -0700 Received: from osmtp1.electric.net ([216.129.90.28]) by zhora.electric.net (NAVGW 2.5.2.12) with SMTP id M2002101014383815721 for ; Thu, 10 Oct 2002 14:38:38 -0700 Received: from [216.210.192.146] (helo=DAVIDFW2K) by osmtp1.electric.net with asmtp (Exim 3.22 #3) id 17zl0w-0000Pe-04 for freebsd-smp@FreeBSD.org; Thu, 10 Oct 2002 14:38:38 -0700 Reply-To: From: "David Francheski" To: Subject: kernel profiling (gprof) on SMP kernels working? Date: Thu, 10 Oct 2002 14:38:36 -0700 Organization: Caymas Systems Message-ID: <002801c270a5$64e1ce70$3600010a@caymas.com> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.3416 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Importance: Normal Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Can anybody tell me if kernel profiling using gprof works reliably (i.e., meaningfully) on SMP kernels? If not, are there any other techniques that can be used to profile an SMP kernel (4.6.2 or 5.0 based)? Thanks! David L. Francheski To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Oct 10 14:47:17 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 700E137B401 for ; Thu, 10 Oct 2002 14:47:16 -0700 (PDT) Received: from mta5.rcsntx.swbell.net (mta5.rcsntx.swbell.net [151.164.30.29]) by mx1.FreeBSD.org (Postfix) with ESMTP id C853843ED4 for ; Thu, 10 Oct 2002 14:47:15 -0700 (PDT) (envelope-from sk@dreamtime.net) Received: from useriwkuwos7hm ([67.114.254.30]) by mta5.rcsntx.swbell.net (iPlanet Messaging Server 5.1 (built May 7 2001)) with ESMTP id <0H3S007EUCIQYM@mta5.rcsntx.swbell.net> for freebsd-smp@FreeBSD.ORG; Thu, 10 Oct 2002 16:47:14 -0500 (CDT) Date: Thu, 10 Oct 2002 14:47:15 -0700 From: Stephen Karrington Subject: SMP kernels working? In-reply-to: <002801c270a5$64e1ce70$3600010a@caymas.com> To: davidf@caymas.com, freebsd-smp@FreeBSD.ORG Message-id: <004e01c270a6$994da5c0$967ba8c0@useriwkuwos7hm> MIME-version: 1.0 X-MIMEOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 X-Mailer: Microsoft Outlook, Build 10.0.2627 Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7BIT Importance: Normal X-Priority: 3 (Normal) X-MSMail-priority: Normal Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I would like to know if anyone has got smp enabled on an Intel motherboard? Thanks. Sincerely, Stephen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Oct 10 22:54:56 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A92C537B401 for ; Thu, 10 Oct 2002 22:54:54 -0700 (PDT) Received: from mx1.freebsd.org (66-178-46-21.reverse.newskies.net [66.178.46.21]) by mx1.FreeBSD.org (Postfix) with SMTP id 1BEC143EB7 for ; Thu, 10 Oct 2002 22:54:52 -0700 (PDT) (envelope-from jim_bankol@surimail.com) From: "Mr. Jim Bankol" Date: Fri, 11 Oct 2002 06:53:52 To: freebsd-smp@freebsd.org Subject: CONFIDENTIAL ASSISTANCE MIME-Version: 1.0 Content-Type: text/plain;charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <20021011055452.1BEC143EB7@mx1.FreeBSD.org> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Dear Sir, First, I must solicit your strictest confidence in this transaction, this is by it's nature of being utterly confidential & top secret. I got your contact through the World Trade Encyclopedia. I am a Sierra Leonean businessman. I was formally into the importation and sale of medical equipment in my native country, Sierra -Leone. My operational base was in Freetown, the capital city. In January 1999, when the second civil war broke out in my country, my office and warehouse were burnt by the Revolutionary United Front Rebels, who invaded Freetown. In that attack, I lost all my properties and assets. I was left with virtually nothing, except When the United Nations sent in troops to quell the rebel's onslaught and there was a relative peace in Freetown, I returned to business, but this time, I invested my resources in the Mining of Diamond which abounds in my country and I exported the Diamond through the help of a third party, some Lebanese who were very much versed in the industry to Antwerp in Belgium. The returns I got from this business was quite good and my life was gradually picking up again until early this year when the Sierra- Leonea. The U.N obliged the government and it is now illegal to sell/buy Diamond emanating from Sierra-Leone. Following this action, the government of my country clamped down on all businessmen who had made fortunes selling Diamond. Consequently, my newly acquired properties were seized again and I had no option than to leave the country immediately. But I did that only after I was able to first move my money out of the country. I had US$10.500,000(Ten Million Five Hundred Thousand United States Dollars) in my domiciliary account in Freetown. I had to move the money out of the Country through a very secret arranged channel. This was done in very unconventional methods, by my friend an official at the "Security Minting & Printing Company" of my country. Now, the money is in OVERSEA COUNTRY under the custody of a private financial outfit in abroad. The help I am now seeking is someone who has the capabilities of receiving the US$10.500,000 in his bank account, because I am on self-exile in a neighboring West African country from where I am trying to work out my traveling documents to enable me travel abroad. All you need to have is a power of attorney from me, empowering you as the recipient of this fund in your bank account. Secondly, after the said fund has been remitted into your bank account , I would want you to help me get a permanent resident permit in your country. For your efforts and the expenses you may incur in the course of this transaction, you will be entitled to 30% of the total sum. But we shall have an agreement to the effect that you WILL NOT take or spend out of my (70%) money until such a time when I am able to be with you. Please, feel free to reach me on my e-mail address (jim_bankol@skarum.zzn.com). When sending an email do remember to include your full name and your direct telephone/fax numbers preferably your private telephone and fax lines. I must apologize for the inconveniences this may cause you. Waiting to hear from you . Best regards, Mr. Jim Bankol. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Oct 10 23:53:43 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1A6A637B401 for ; Thu, 10 Oct 2002 23:53:42 -0700 (PDT) Received: from gw.one.com.au (gw.one.com.au [203.18.85.1]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1CD8743E7B for ; Thu, 10 Oct 2002 23:53:33 -0700 (PDT) (envelope-from raymond@one.com.au) Received: from one.com.au (pmo.local [10.18.85.2]) by gw.one.com.au (8.12.2/8.12.2) with SMTP id g9B6rI06025816 for smp@freebsd.org; Fri, 11 Oct 2002 16:53:24 +1000 (EST) (envelope-from raymond@one.com.au) Date: Fri, 11 Oct 2002 16:53:18 +1000 (EST) From: User Raymond Message-Id: <200210110653.g9B6rI06025816@gw.one.com.au> Subject: SMP problems with MSI MB and dual Xeon Subj: SMP problems with MSI MB and dual Xeon To: smp@freebsd.org X-Spam-Score: -96.4 () DATE_MISSING,INVALID_DATE,DOUBLE_CAPSWORD,USER_IN_WHITELIST X-Scanned-By: MIMEDefang 2.21 (www . roaringpenguin . com / mimedefang) Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org The following problem with an MSI E7500 MB with dual 2.0GHZ Xeon processors looks similar to: Problem Report i386/40564 Problem Report misc/42414 Symptoms are - build SMP kernel and boot - output ends at: Changing APIC ID for IO APIC #3 from 7 to 15 in MP table Changing APIC ID for IO APIC #4 from 15 to 8 on chip panic: can't control APIC #4 ID, reg: 0xffffffff mp_lock = 00000001; cpuid = 0; lapic.id = 00000000 The MSI support people suggest the following: > You need to install patch for Unix. > So I would suggest you to look on FreeBSD's website for patch supporting > Intel Plumas chipset. and > Also if the customer is using Xeon with HyperThread function, they may need > to install the patch for that as well. > For the support of Prestonia processors. I have no idea as to what they are refering to... Any suggestions would be appreciated. Ray Newman 11 Oct 2002 Message sent at 04:52 PM on 11 Oct 2002 by PUP::RAYMOND. Id: 191855. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Oct 11 0:37:53 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id DE45537B401 for ; Fri, 11 Oct 2002 00:37:51 -0700 (PDT) Received: from mx.estimese.net (finom.estimese.net [195.168.3.50]) by mx1.FreeBSD.org (Postfix) with SMTP id 8F64843EB2 for ; Fri, 11 Oct 2002 00:37:50 -0700 (PDT) (envelope-from zero@estimese.net) Received: (qmail 10433 invoked by uid 69); 11 Oct 2002 07:49:21 -0000 Date: Fri, 11 Oct 2002 09:49:21 +0200 From: Robert Bopko To: User Raymond Cc: smp@freebsd.org Subject: Re: SMP problems with MSI MB and dual Xeon Message-ID: <20021011074921.GD10226@finom.estimese.net> References: <200210110653.g9B6rI06025816@gw.one.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200210110653.g9B6rI06025816@gw.one.com.au> User-Agent: Mutt/1.4i X-Operating-System: NetBSD/i386 Organization: Nextra Slovakia, Ltd. Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org hello, it looks like motherboards with E7500 chipset are not good friends with freebsd-smp. can someone pls tell me why? thanks. On Fri, Oct 11, 2002 at 04:53:18PM +1000, User Raymond wrote: > > The following problem with an MSI E7500 MB with dual 2.0GHZ Xeon processors > looks similar to: > > Problem Report i386/40564 > Problem Report misc/42414 > > Symptoms are - build SMP kernel and boot - output ends at: > > Changing APIC ID for IO APIC #3 from 7 to 15 in MP table > Changing APIC ID for IO APIC #4 from 15 to 8 on chip > panic: can't control APIC #4 ID, reg: 0xffffffff > mp_lock = 00000001; cpuid = 0; lapic.id = 00000000 > > > The MSI support people suggest the following: > > > You need to install patch for Unix. > > So I would suggest you to look on FreeBSD's website for patch supporting > > Intel Plumas chipset. > > and > > > Also if the customer is using Xeon with HyperThread function, they may need > > to install the patch for that as well. > > For the support of Prestonia processors. > > I have no idea as to what they are refering to... > > Any suggestions would be appreciated. > > > Ray Newman > 11 Oct 2002 > Message sent at 04:52 PM on 11 Oct 2002 by PUP::RAYMOND. Id: 191855. > > 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 Fri Oct 11 3:47:50 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id DE24037B401 for ; Fri, 11 Oct 2002 03:47:49 -0700 (PDT) Received: from mailman.zeta.org.au (mailman.zeta.org.au [203.26.10.16]) by mx1.FreeBSD.org (Postfix) with ESMTP id E86FF43EAA for ; Fri, 11 Oct 2002 03:47:48 -0700 (PDT) (envelope-from bde@zeta.org.au) Received: from bde.zeta.org.au (bde.zeta.org.au [203.2.228.102]) by mailman.zeta.org.au (8.9.3/8.8.7) with ESMTP id UAA24699; Fri, 11 Oct 2002 20:47:36 +1000 Date: Fri, 11 Oct 2002 20:57:49 +1000 (EST) From: Bruce Evans X-X-Sender: bde@gamplex.bde.org To: David Francheski Cc: freebsd-smp@FreeBSD.ORG Subject: Re: kernel profiling (gprof) on SMP kernels working? In-Reply-To: <002801c270a5$64e1ce70$3600010a@caymas.com> Message-ID: <20021011204219.C12512-100000@gamplex.bde.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Thu, 10 Oct 2002, David Francheski wrote: > Can anybody tell me if kernel profiling using gprof > works reliably (i.e., meaningfully) on SMP kernels? It worked not very meaningfully in -current last time I tried it (about 6 months ago; I only run -current, rarely run SMP, and sometimes run profiling). It makes no attempt to separate the CPUs. This works OK for the flat profile, except you can't see if the CPUs are sharing the work evenly. It's not clear that the algorithms even give reasonable results for the call graph, but the results seemed reasonable. I didn't check them very carefully (I normally only look at the flat profile except for interesting functions). > If not, are there any other techniques that can be > used to profile an SMP kernel (4.6.2 or 5.0 based)? None that I know of which do similar things or do them better. The MUTEX_PROFILING option in -current will tell you that -current spends too much time waiting for locks, and where. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Oct 11 12:14: 7 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9D2C237B401 for ; Fri, 11 Oct 2002 12:14:06 -0700 (PDT) Received: from posgate.acis.com.au (posgate.acis.com.au [203.14.230.14]) by mx1.FreeBSD.org (Postfix) with ESMTP id 08F1143E9E for ; Fri, 11 Oct 2002 12:14:05 -0700 (PDT) (envelope-from andymac@bullseye.apana.org.au) Received: from bullseye.apana.org.au (dialup-1.acis.com.au [203.14.230.80] (may be forged)) by posgate.acis.com.au (8.11.6/8.11.6) with ESMTP id g9BJE1029041 for ; Sat, 12 Oct 2002 05:14:02 +1000 Received: from bullseye.apana.org.au (tenring.andymac.org [203.9.107.238]) by bullseye.apana.org.au (8.11.6/8.11.6) with ESMTP id g9BBGiI42758 for ; Fri, 11 Oct 2002 21:16:44 +1000 (EST) (envelope-from andymac@bullseye.apana.org.au) Date: Fri, 11 Oct 2002 21:15:18 +1100 (edt) From: Andrew MacIntyre To: Subject: Re: Abit SMP boards under FreeBSD (was Re: dnetc on xeon) In-Reply-To: <20021010100823.0a1e00ad.michaelossareh@onetel.net.uk> Message-ID: X-X-Sender: andymac@bullseye.apana.org.au MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Thu, 10 Oct 2002, Michael Ossareh wrote: > I believe the Abit VP6 and BP6 are quite similar and the chipset runs > very hot when in SMP mode, so since its not a proper server board you > might want to invest in some active cooling for your chipset (usually > just has a green heat sink on it). I think the VP6 uses the VIA [56]94 chipset, whereas the BP6 uses the Intel 440BX chipset - 2 very different animals. I remember some mail about problems with SMP with that VIA chipset, which I kept for a while but deleted only recently. -- Andrew I MacIntyre "These thoughts are mine alone..." E-mail: andymac@bullseye.apana.org.au | Snail: PO Box 370 andymac@pcug.org.au | Belconnen ACT 2616 Web: http://www.andymac.org/ | Australia To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Oct 11 15:50: 1 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 6834137B401 for ; Fri, 11 Oct 2002 15:50:00 -0700 (PDT) Received: from swordfish.cs.caltech.edu (swordfish.cs.caltech.edu [131.215.44.124]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1662143E6A for ; Fri, 11 Oct 2002 15:50:00 -0700 (PDT) (envelope-from chengjin@cs.caltech.edu) Received: from fast2.cs.caltech.edu (fast2.cs.caltech.edu [131.215.45.55]) by swordfish.cs.caltech.edu (Postfix) with ESMTP id 43D72DF260; Fri, 11 Oct 2002 15:49:51 -0700 (PDT) Received: from localhost (chengjin@localhost) by fast2.cs.caltech.edu (8.11.6/8.9.3) with ESMTP id g9BMnpl31160; Fri, 11 Oct 2002 15:49:51 -0700 X-Authentication-Warning: fast2.cs.caltech.edu: chengjin owned process doing -bs Date: Fri, 11 Oct 2002 15:49:50 -0700 (PDT) From: Cheng Jin To: Cc: Xiaoliang Wei Subject: SMP + dual GbE Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Hi, We are currently working on a high-speed TCP project. We have a couple of machines built on supermicro's P4DP6 board each with two 2.4 Ghz Xeon processors and two GbE cards from SysKonnect. We were able to get close to 1 Gbps throughput(loopback tests) on RedHat Linux 7.3 SMP kernel (2.4.18-3) with interrupt moderation on the GbE cards (5000 interrupts/s) and binding one CPU to each card. However, under FreeBSD SMP, we have only been able to achieve about 640 Mbps on each card. I suppose the binding of CPU to card is implicit (?) under FreeBSD SMP since we didn't notice any out-of-order packets. We tried interrupt moderation going as high as 10,000 per second and as low as 2500, but that didn't seem to help much. Furthermore, we were able to get about 560 Mbps throughput for each card under a single-processor kernel so it seems that the extra processor didn't help much under SMP. We don't know a whole lot about how SMP works under FreeBSD. Actually, we don't know anything about FreeBSD SMP. Any comments/suggestions are greatly appreciated. Cheng Lab # 626 395 8820 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Oct 11 17: 1:50 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BDB4637B401 for ; Fri, 11 Oct 2002 17:01:48 -0700 (PDT) Received: from snipe.mail.pas.earthlink.net (snipe.mail.pas.earthlink.net [207.217.120.62]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2B2A043EAA for ; Fri, 11 Oct 2002 17:01:08 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from pool0285.cvx40-bradley.dialup.earthlink.net ([216.244.43.30] helo=mindspring.com) by snipe.mail.pas.earthlink.net with esmtp (Exim 3.33 #1) id 1809hi-0000h3-00; Fri, 11 Oct 2002 17:00:27 -0700 Message-ID: <3DA765D1.D6C4B92C@mindspring.com> Date: Fri, 11 Oct 2002 16:59:13 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Cheng Jin Cc: freebsd-smp@freebsd.org, Xiaoliang Wei Subject: Re: SMP + dual GbE References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Cheng Jin wrote: > We are currently working on a high-speed TCP project. We have a couple of > machines built on supermicro's P4DP6 board each with two 2.4 Ghz Xeon > processors and two GbE cards from SysKonnect. We were able to get close to > 1 Gbps throughput(loopback tests) on RedHat Linux 7.3 SMP kernel > (2.4.18-3) with interrupt moderation on the GbE cards (5000 interrupts/s) > and binding one CPU to each card. > > However, under FreeBSD SMP, we have only been able to achieve about 640 > Mbps on each card. I suppose the binding of CPU to card is implicit (?) > under FreeBSD SMP since we didn't notice any out-of-order packets. We > tried interrupt moderation going as high as 10,000 per second and as low > as 2500, but that didn't seem to help much. Furthermore, we were able to get > about 560 Mbps throughput for each card under a single-processor kernel so > it seems that the extra processor didn't help much under SMP. > > We don't know a whole lot about how SMP works under FreeBSD. Actually, we > don't know anything about FreeBSD SMP. Any comments/suggestions are > greatly appreciated. It may be your SysKonnect card. It's very easy to get full wire speed with 1500 MTU packets on an 800MHz machine with Tigon III cards (for example). It's all in the tuning. See previous threads in the FreeBSD-net mailing list archives for more information. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Oct 11 17:13:16 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1359337B41B for ; Fri, 11 Oct 2002 17:13:14 -0700 (PDT) Received: from swordfish.cs.caltech.edu (swordfish.cs.caltech.edu [131.215.44.124]) by mx1.FreeBSD.org (Postfix) with ESMTP id AFD7A43E42 for ; Fri, 11 Oct 2002 17:13:13 -0700 (PDT) (envelope-from chengjin@cs.caltech.edu) Received: from fast2.cs.caltech.edu (fast2.cs.caltech.edu [131.215.45.55]) by swordfish.cs.caltech.edu (Postfix) with ESMTP id E3966DF260; Fri, 11 Oct 2002 17:13:12 -0700 (PDT) Received: from localhost (chengjin@localhost) by fast2.cs.caltech.edu (8.11.6/8.9.3) with ESMTP id g9C0DCX31594; Fri, 11 Oct 2002 17:13:12 -0700 X-Authentication-Warning: fast2.cs.caltech.edu: chengjin owned process doing -bs Date: Fri, 11 Oct 2002 17:13:12 -0700 (PDT) From: Cheng Jin To: Terry Lambert Cc: "freebsd-smp@freebsd.org" , Xiaoliang Wei Subject: Re: SMP + dual GbE In-Reply-To: <3DA765D1.D6C4B92C@mindspring.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org > It may be your SysKonnect card. It's very easy to get full wire > speed with 1500 MTU packets on an 800MHz machine with Tigon III > cards (for example). It's all in the tuning. Sorry, I didn't make myself clear in my previous message. We were able to get 1 Gbps when we pump data through one SysKonnect card (SMP or not). The problem is when we try to use both cards under FreeBSD SMP kernel, the throughput on each card is only 640 Mbps. In fact, using a single-processor kernel, we could get 560 Mbps going through both cards. The cards seem fine, the problem seems to be with the SMP kernel. Thanks, Cheng To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Oct 11 17:26: 2 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 70EAC37B401 for ; Fri, 11 Oct 2002 17:26:01 -0700 (PDT) Received: from gull.mail.pas.earthlink.net (gull.mail.pas.earthlink.net [207.217.120.84]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0F4C343EAF for ; Fri, 11 Oct 2002 17:26:01 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from pool0285.cvx40-bradley.dialup.earthlink.net ([216.244.43.30] helo=mindspring.com) by gull.mail.pas.earthlink.net with esmtp (Exim 3.33 #1) id 180A6R-000667-00; Fri, 11 Oct 2002 17:26:00 -0700 Message-ID: <3DA76BCA.B4EF9D79@mindspring.com> Date: Fri, 11 Oct 2002 17:24:42 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Cheng Jin Cc: "freebsd-smp@freebsd.org" , Xiaoliang Wei Subject: Re: SMP + dual GbE References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Cheng Jin wrote: > > It may be your SysKonnect card. It's very easy to get full wire > > speed with 1500 MTU packets on an 800MHz machine with Tigon III > > cards (for example). It's all in the tuning. > > Sorry, I didn't make myself clear in my previous message. We were able to > get 1 Gbps when we pump data through one SysKonnect card (SMP or not). > The problem is when we try to use both cards under FreeBSD SMP kernel, the > throughput on each card is only 640 Mbps. In fact, using a > single-processor kernel, we could get 560 Mbps going through both cards. > > The cards seem fine, the problem seems to be with the SMP kernel. FreeBSD never routes interrupts to more than a single processor at a time. I believe Linux does the same, though it does not select the CPU to target the same way. NT assigns cards to CPUs to distribute the load. As things currently stand, you should always get better overall numbers on tests like this with a UP FreeBSD system, unless you are willing to take suggestions on your test configuration. 33Mhz x 32 bits ~= 1.32Gbit burst rate; this could be a PCI bus limitation. You didn't say how fast your PCI bus was, whether you were using 64 bit slots for each card, whether they were/were not sharing a PCI interrupt with other cards, etc.. You might also consider contacting the driver's author, since he would best know how to get good numbers out of the hardware. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Oct 11 20:48:21 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 789BB37B401 for ; Fri, 11 Oct 2002 20:48:20 -0700 (PDT) Received: from absinthe.carnagecopia.com (absinthe.carnagecopia.com [216.187.87.246]) by mx1.FreeBSD.org (Postfix) with SMTP id 0168B43E42 for ; Fri, 11 Oct 2002 20:48:16 -0700 (PDT) (envelope-from random@goblinstudios.com) Received: (qmail 91569 invoked by uid 85); 12 Oct 2002 03:48:05 -0000 Received: from random@goblinstudios.com by absinthe.carnagecopia.com with qmail-scanner-1.03 (uvscan: v4.1.60/v4228. . Clean. Processed in 0.392823 secs); 12 Oct 2002 03:48:05 -0000 Received: from unknown (HELO workstation-22.internal.carnagecopia.com) (204.244.192.2) by absinthe.carnagecopia.com with SMTP; 12 Oct 2002 03:48:05 -0000 Date: Fri, 11 Oct 2002 23:48:49 -0400 From: Vincent Janelle To: Terry Lambert Cc: chengjin@cs.caltech.edu, freebsd-smp@freebsd.org, weixl@cs.caltech.edu Subject: Re: SMP + dual GbE Message-Id: <20021011234849.0644d5ba.random@goblinstudios.com> In-Reply-To: <3DA76BCA.B4EF9D79@mindspring.com> References: <3DA76BCA.B4EF9D79@mindspring.com> Organization: http://www.goblinstudios.com/ X-Mailer: Sylpheed version 0.8.1 (GTK+ 1.2.10; i586-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org The other thing to check for is that you are not putting 32bit or 33Mhz cards in the 64bit/66Mhz slots. This will force down the overall speed down the slowest device.. And also reduce the bus speed between the north and south chipsets (depending of course on what chipset you are using). On Fri, 11 Oct 2002 17:24:42 -0700 Terry Lambert wrote: > Cheng Jin wrote: > > > It may be your SysKonnect card. It's very easy to get full wire > > > speed with 1500 MTU packets on an 800MHz machine with Tigon III > > > cards (for example). It's all in the tuning. > > > > Sorry, I didn't make myself clear in my previous message. We were able to > > get 1 Gbps when we pump data through one SysKonnect card (SMP or not). > > The problem is when we try to use both cards under FreeBSD SMP kernel, the > > throughput on each card is only 640 Mbps. In fact, using a > > single-processor kernel, we could get 560 Mbps going through both cards. > > > > The cards seem fine, the problem seems to be with the SMP kernel. > > FreeBSD never routes interrupts to more than a single processor > at a time. I believe Linux does the same, though it does not > select the CPU to target the same way. NT assigns cards to CPUs > to distribute the load. > > As things currently stand, you should always get better overall > numbers on tests like this with a UP FreeBSD system, unless you > are willing to take suggestions on your test configuration. > > 33Mhz x 32 bits ~= 1.32Gbit burst rate; this could be a PCI bus > limitation. > > You didn't say how fast your PCI bus was, whether you were using > 64 bit slots for each card, whether they were/were not sharing a > PCI interrupt with other cards, etc.. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Oct 12 11: 1:45 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 87BC137B401 for ; Sat, 12 Oct 2002 11:01:42 -0700 (PDT) Received: from brodie.zefram.net (dsl092-065-196.bos1.dsl.speakeasy.net [66.92.65.196]) by mx1.FreeBSD.org (Postfix) with SMTP id 40D2443E97 for ; Sat, 12 Oct 2002 11:01:41 -0700 (PDT) (envelope-from zefram@zefram.net) Received: (qmail 6658 invoked by uid 1000); 12 Oct 2002 18:01:32 -0000 Received: from localhost (sendmail-bs@127.0.0.1) by localhost with SMTP; 12 Oct 2002 18:01:32 -0000 Date: Sat, 12 Oct 2002 14:01:31 -0400 (EDT) From: John Gillis To: freebsd-smp@freebsd.org Subject: MP Tables Hosed... huh? Message-ID: <20021012134727.Q6501-100000@brodie.zefram.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I recently installed FreeBSD 4.7 on a Compaq DL580 server with dual Xeon 700s. Everything went fine until it was time to recompile the kernel with SMP support. Instead of getting a nice error message telling me what was wrong, the system just hangs on the bootup message below. When I try to run MPTable, it reports the extended tables are hosed. Not sure what that means. I've included the bootup error message (I'd give the whole bootup message but it's gone when I reboot) and the MPTable output. If anyone could offer any advice on how to get SMP enabled, I would appreciate it. Thank you for your time, John ------------------------------------------------------------ Bootup error message: Changing APIC ID for IO APIC #0 from 0 to 8 on chip Programming 35 pins in IOAPIC #0 IOAPIC #0 intpin 2 -> irq 0 ------------------------------------------------------------ MPTable error output: =============================================================================== MPTable, version 2.0.15 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000f4fd0 signature: '_MP_' length: 16 bytes version: 1.4 checksum: 0x00 mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x000f2f1e signature: 'PCMP' base table length: 492 version: 1.4 checksum: 0xd0 OEM ID: 'COMPAQ ' Product ID: 'PROLIANT ' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 53 local APIC address: 0xfee00000 extended table length: 172 extended table checksum: 241 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 0 0x10 BSP, usable 6 2 1 0x0381 2 0x10 AP, usable 6 10 1 0x383fbff -- Bus: Bus ID Type 0 PCI 1 PCI 2 PCI 15 ISA -- I/O APICs: APIC ID Version State Address 8 0x11 usable 0xfec00000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# INT active-lo level 1 6:A 8 16 INT active-lo level 1 6:B 8 17 INT active-lo level 1 6:C 8 16 INT active-lo level 1 6:D 8 17 INT active-lo level 1 8:A 8 18 INT active-lo level 1 8:B 8 19 INT active-lo level 1 8:C 8 18 INT active-lo level 1 8:D 8 19 INT active-lo level 1 9:A 8 20 INT active-lo level 1 9:B 8 21 INT active-lo level 1 9:C 8 20 INT active-lo level 1 9:D 8 21 INT active-lo level 1 5:A 8 30 INT active-lo level 2 5:A 8 30 INT active-lo level 2 6:A 8 22 INT active-lo level 2 6:B 8 23 INT active-lo level 2 6:C 8 22 INT active-lo level 2 6:D 8 23 INT active-lo level 2 7:A 8 24 INT active-lo level 2 7:B 8 25 INT active-lo level 2 7:C 8 24 INT active-lo level 2 7:D 8 25 INT active-lo level 0 7:A 8 26 INT active-lo level 0 7:B 8 27 INT active-lo level 0 7:C 8 26 INT active-lo level 0 7:D 8 27 INT active-lo level 0 4:A 8 29 INT active-lo level 0 4:B 8 28 INT active-hi edge 15 1 8 1 INT active-hi edge 15 0 8 2 INT active-hi edge 15 3 8 3 INT active-hi edge 15 4 8 4 INT active-hi edge 15 5 8 5 INT active-hi edge 15 6 8 6 INT active-hi edge 15 7 8 7 INT active-hi edge 15 8 8 8 INT active-hi edge 15 9 8 9 INT active-hi edge 15 10 8 10 INT active-hi edge 15 11 8 11 INT active-hi edge 15 12 8 12 INT active-lo level 15 13 8 13 INT active-hi edge 15 14 8 14 INT active-hi edge 15 15 8 15 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT conforms conforms 15 0 255 0 NMI conforms conforms 15 0 255 1 ------------------------------------------------------------------------------- MP Config Extended Table Entries: Extended Table HOSED! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message