From owner-freebsd-smp Sun Aug 31 12:09:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id MAA13980 for smp-outgoing; Sun, 31 Aug 1997 12:09:13 -0700 (PDT) Received: from tweetie-bird.cs.washington.edu (tweetie-bird.cs.washington.edu [128.95.2.46]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id MAA13972 for ; Sun, 31 Aug 1997 12:09:09 -0700 (PDT) From: mef@cs.washington.edu Received: (mef@localhost) by tweetie-bird.cs.washington.edu (8.8.5+CS/7.2ws+) id MAA12729; Sun, 31 Aug 1997 12:09:24 -0700 (PDT) Date: Sun, 31 Aug 1997 12:09:24 -0700 (PDT) Message-Id: <199708311909.MAA12729@tweetie-bird.cs.washington.edu> To: Shimon@i-connect.net CC: FreeBSD-SMP@FreeBSD.ORG In-reply-to: (message from Simon Shapiro on Fri, 15 Aug 1997 22:55:50 -0700 (PDT)) Subject: Re: Open Issues on P6DNH Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Simon, What the deal with the P6DNH motherboard from Supermicro? A while back you posted a set of problems to the smp mailing list. Are these now resolved? Is the P6DNH MB a good one to buy if one wants to play around with i2o? Thanks, Marc Date: Fri, 15 Aug 1997 22:55:50 -0700 (PDT) Organization: Atlas Telecom From: Simon Shapiro Hi Steve, Sorry for the long delay. Had to design and code a new filesystem type, fight a bunch of political battles, and get a misely raise after a wonderful review (``words are cheaper than cash'', was my comment). Hope you are doing well. * Although it compiles, the SoundBlaster driver is less than functional; cat foo.au > /dev/audio just hangs there. Works fine under UP on the same platform. * The fxp driver seems to not get interrupts at all. Everything times out eventually and works like a very slow polled system. This happens IN FRONT of the i960 bridge. The fpa (fddi) driver, the DPT driver all work fine. * If you put COM4 on irq 7 it is there, on irq 9 it is not found. A conflict with the sb? The sb is mapped at 220. com4 is 2e8. * Something strange. Booted SMP, started X11, xperfmon++ insisted that libXaw was not there if ``shimon'' tried to tun it. Was there if root ran it. Permissions? Nope. I re-ompiled and re-installed it. Now root could see the executable (in /usr/X11R6/bin) but not ``shimon''. Rebooted and both are fine. Never seen this one before. * Interrupt latency is still not so hot. The typical DPT driver in UP serves an interrupt in 3-55us. in SMP it takes 11-1000039us. * I am getting an enormous number of spurious interrupts. In UP the typical count is 0 aborted (due to errors) and 1 for spurious (int routine was called but hardware said ``not me''. In SMP, the numbrs are 0 and 62090. The 1 spurious is explainable ( a bug in transitioning from polled mode to interrupt mode during boot. If there is anything I can do to help, let me know. I have sendero-ppp now run smp. I have another machine I will setup tomorrow (new Brass) which will be the same hardware. Simon From owner-freebsd-smp Sun Aug 31 13:00:21 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id NAA15845 for smp-outgoing; Sun, 31 Aug 1997 13:00:21 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id NAA15837 for ; Sun, 31 Aug 1997 13:00:17 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id OAA21270; Sun, 31 Aug 1997 14:00:11 -0600 (MDT) Message-Id: <199708312000.OAA21270@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: mef@cs.washington.edu cc: Shimon@i-connect.net, FreeBSD-SMP@FreeBSD.ORG Subject: Re: Open Issues on P6DNH In-reply-to: Your message of "Sun, 31 Aug 1997 12:09:24 PDT." <199708311909.MAA12729@tweetie-bird.cs.washington.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 31 Aug 1997 14:00:10 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > Simon, > > What the deal with the P6DNH motherboard from Supermicro? A while > back you posted a set of problems to the smp mailing list. Are these > now resolved? Is the P6DNH MB a good one to buy if one wants to play > around with i2o? There is a problem with the mptable provided by the BIOS. Until thats fixed or I get more info from supermicro about whats going on, its a bad choice for freebsd SMP. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Sun Aug 31 13:10:38 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id NAA16312 for smp-outgoing; Sun, 31 Aug 1997 13:10:38 -0700 (PDT) Received: from ic.net (qmailr@srv1b.ic.net [152.160.72.20]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id NAA16307 for ; Sun, 31 Aug 1997 13:10:36 -0700 (PDT) Received: (qmail 21888 invoked from network); 31 Aug 1997 20:10:20 -0000 Received: from unknown (HELO lurch.rickl.org) (152.160.108.28) by unknown with SMTP; 31 Aug 1997 20:10:20 -0000 Received: from lurch.rickl.org (rickl@lurch.rickl.org [192.168.255.1]) by lurch.rickl.org (8.8.7/8.7.3) with ESMTP id QAA00330 for ; Sun, 31 Aug 1997 16:09:33 -0400 (EDT) Date: Sun, 31 Aug 1997 16:09:33 -0400 (EDT) From: Rick Lotoczky Reply-To: Rick Lotoczky Subject: SMP & P6DOF - Show of Hands To: smp@freebsd.org Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi Steve, Sorry for the late reply, but I was out of town for the last week. Seems like the changes you made to SMP system were a big win here. Silo overflows are non-existant, a few interrupt-overflows remain. I can now access the IDE drives **and** connect with my 2B ISDN line concurrently. In the past, the sio would all but stop during a disk transfer. The VM stuff made quite a difference also. I can now compile the whole system usinf SMP without hanging. A week or so back, the only way to compile the system was to use the uni-processor kernel. Great job to all!! Rick From owner-freebsd-smp Sun Aug 31 13:12:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id NAA16393 for smp-outgoing; Sun, 31 Aug 1997 13:12:12 -0700 (PDT) Received: from neutron.neutron.org (neutron.neutron.org [207.167.87.103]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id NAA16386 for ; Sun, 31 Aug 1997 13:12:08 -0700 (PDT) Received: (from smp@localhost) by neutron.neutron.org (8.8.5/8.8.5) id NAA00844 for freebsd-smp@freebsd.org; Sun, 31 Aug 1997 13:12:33 -0700 (PDT) Date: Sun, 31 Aug 1997 13:12:33 -0700 (PDT) From: smp Message-Id: <199708312012.NAA00844@neutron.neutron.org> To: freebsd-smp@freebsd.org Subject: motherboard choice Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk hello i am interested in getting into SMP. i need to pick a good motherboard. the intel PR440FX (100 M/s ethernet, adaptec UW scsi) with 2 200 MHZ PPros can be had for abt $1.5K. the tyan tomcat IV can handle 2 233 MHZ parts, but no scsi or ethernet. or any other suggestions? comments, or advice before i buy? thanks bill clarke dept of physics, u.c.s.c. (smp@neutron.org) From owner-freebsd-smp Sun Aug 31 13:18:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id NAA16688 for smp-outgoing; Sun, 31 Aug 1997 13:18:06 -0700 (PDT) Received: from neutron.neutron.org (neutron.neutron.org [207.167.87.103]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id NAA16683 for ; Sun, 31 Aug 1997 13:18:02 -0700 (PDT) Received: (from smp@localhost) by neutron.neutron.org (8.8.5/8.8.5) id NAA00837 for FreeBSD-SMP@FreeBSD.ORG; Sun, 31 Aug 1997 13:09:02 -0700 (PDT) Date: Sun, 31 Aug 1997 13:09:02 -0700 (PDT) From: smp Message-Id: <199708312009.NAA00837@neutron.neutron.org> To: FreeBSD-SMP@FreeBSD.ORG Subject: motherboard choice Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk hello i am interested in getting into SMP. i need to pick a good motherboard. the intel PR440FX (100 M/s ethernet, adaptec UW scsi) with 2 200 MHZ PPros can be had for abt $1.5K. the tyan tomcat IV can handle 2 233 MHZ parts, but no scsi or etthernet. or any other suggestions? comments, or advice before i buy? thanks bill clarke dept of physics, u.c.s.c. (smp@neutron.org) From owner-freebsd-smp Sun Aug 31 14:16:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA18768 for smp-outgoing; Sun, 31 Aug 1997 14:16:04 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA18763 for ; Sun, 31 Aug 1997 14:16:01 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id PAA21499; Sun, 31 Aug 1997 15:14:41 -0600 (MDT) Message-Id: <199708312114.PAA21499@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: smp cc: FreeBSD-SMP@FreeBSD.ORG Subject: Re: motherboard choice In-reply-to: Your message of "Sun, 31 Aug 1997 13:09:02 PDT." <199708312009.NAA00837@neutron.neutron.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 31 Aug 1997 15:14:41 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > i am interested in getting into SMP. i need to pick a good motherboard. > the intel PR440FX (100 M/s ethernet, adaptec UW scsi) with 2 200 MHZ PPros > can be had for abt $1.5K. > the tyan tomcat IV can handle 2 233 MHZ parts, but no scsi or etthernet. > or any other suggestions? comments, or advice before i buy? > thanks The PR440 can also do 233mHz, the jumpers are there, just undocumented. The big negative of the intel is that it only has 1 ISA slot. I personally suggest the gigabyte ga686dx over the tyan. Iwill makes a dual P6 with the adaptec UW on it, but I'm not sure if anyone has reported success with it yet. Consider the P6-166x512 chips. I am overclocking mine to 200 (obviously no guarantees about this), and the extra cache is *probably* more beneficial than 33mHz. They can be had for around $320 each. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Sun Aug 31 14:18:15 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA18860 for smp-outgoing; Sun, 31 Aug 1997 14:18:15 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA18855 for ; Sun, 31 Aug 1997 14:18:12 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id PAA21554; Sun, 31 Aug 1997 15:18:10 -0600 (MDT) Message-Id: <199708312118.PAA21554@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Rick Lotoczky cc: smp@FreeBSD.ORG Subject: Re: SMP & P6DOF - Show of Hands In-reply-to: Your message of "Sun, 31 Aug 1997 16:09:33 EDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 31 Aug 1997 15:18:09 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > like the changes you made to SMP system were a big win here. Silo overflows > are non-existant, a few interrupt-overflows remain. I can now access the > IDE drives **and** connect with my 2B ISDN line concurrently. In the past, > the sio would all but stop during a disk transfer. The VM stuff made quite > a difference also. I can now compile the whole system usinf SMP without > hanging. A week or so back, the only way to compile the system was to use > the uni-processor kernel. Great job to all!! I've been making alot of changes that would affect this. could you say exactly when you last updated your code? -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Sun Aug 31 14:19:09 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA18891 for smp-outgoing; Sun, 31 Aug 1997 14:19:09 -0700 (PDT) Received: from news1.gtn.com (news1.gtn.com [194.77.0.15]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA18886 for ; Sun, 31 Aug 1997 14:19:05 -0700 (PDT) Received: (from uucp@localhost) by news1.gtn.com (8.8.6/8.8.6) with UUCP id XAA28399; Sun, 31 Aug 1997 23:15:18 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.7/8.8.7) id XAA01967; Sun, 31 Aug 1997 23:11:30 +0200 (CEST) Message-ID: <19970831231130.52301@klemm.gtn.com> Date: Sun, 31 Aug 1997 23:11:30 +0200 From: Andreas Klemm To: Steve Passe Cc: smp@FreeBSD.ORG Subject: Re: HEADS UP: another set of changes. References: <199708300852.CAA11436@Ilsa.StevesCafe.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79 In-Reply-To: <199708300852.CAA11436@Ilsa.StevesCafe.com>; from Steve Passe on Sat, Aug 30, 1997 at 02:52:52AM -0600 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT SMP Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sat, Aug 30, 1997 at 02:52:52AM -0600, Steve Passe wrote: > > We're at the point where deadlocks are going to become common. Bear with me > and help as you can. When you hit one please record all the facts you can, > and report them to smp@freebsd.org. Hopefully we will get thru this period > before too long... Yesterday evening and today evening my system hangs with a page fault. I'm on CPU#1, I need to be on CPU#0, sleeping timeout waiting for CPU#0! syncing disks... Fatal trap 12: page fault while in kernel mode cpuid=1 lapic.id=0 fault virtual address = 0x10 fault code = supervisor-read, page not present instruction pointer = 0x8: 0xf0133110 stack pointer = 0x10:0xff803c5c frame pointer = 0x10:0xff803c70 code segment = base 0x0, limit 0xfffff, type=0x1b = DPL0, pres 1, def 32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = Idle mp_lock = 01000004 interrupt mask = tty <- SMP: XXX trap number = 12 panic: page fault cpuid 1 This was the output I noticed yestwerday evening. Today I cvsupped newly and build a new kernel. When the panic happened today at about 10pm, I compared the output with that above messages I wrote down, and everything is nearly the same, only the instruction stack and frame pointer were about off by one or two, because of the new kernel I think. What do you need to have more, so that you can figure out, what's going wrong ? Andreas /// -- Andreas Klemm | klemm.gtn.com - powered by Symmetric MultiProcessor FreeBSD http://www.freebsd.org/~fsmp/SMP/SMP.html http://www.freebsd.org/~fsmp/SMP/benches.html From owner-freebsd-smp Sun Aug 31 14:26:47 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA19140 for smp-outgoing; Sun, 31 Aug 1997 14:26:47 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA19135 for ; Sun, 31 Aug 1997 14:26:44 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id PAA21626; Sun, 31 Aug 1997 15:25:59 -0600 (MDT) Message-Id: <199708312125.PAA21626@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Andreas Klemm cc: smp@FreeBSD.ORG Subject: Re: HEADS UP: another set of changes. In-reply-to: Your message of "Sun, 31 Aug 1997 23:11:30 +0200." <19970831231130.52301@klemm.gtn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 31 Aug 1997 15:25:58 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > Yesterday evening and today evening my system hangs with a page fault. > > I'm on CPU#1, I need to be on CPU#0, sleeping > timeout waiting for CPU#0! > syncing disks... > Fatal trap 12: page fault while in kernel mode > cpuid=1 > lapic.id=0 > fault virtual address = 0x10 > fault code = supervisor-read, page not present > instruction pointer = 0x8: 0xf0133110 > stack pointer = 0x10:0xff803c5c > frame pointer = 0x10:0xff803c70 > code segment = base 0x0, limit 0xfffff, type=0x1b > = DPL0, pres 1, def 32 1, gran 1 > processor eflags = interrupt enabled, resume, IOPL = 0 > current process = Idle > mp_lock = 01000004 > interrupt mask = tty <- SMP: XXX > trap number = 12 > panic: page fault > cpuid 1 > > This was the output I noticed yestwerday evening. > Today I cvsupped newly and build a new kernel. > When the panic happened today at about 10pm, I compared the > output with that above messages I wrote down, and everything > is nearly the same, only the instruction stack and frame pointer > were about off by one or two, because of the new kernel I think. > > What do you need to have more, so that you can figure out, what's > going wrong ? we need a trace. also, it looks like it was shutting down when this happened? specifically while syncing disks. had you down a shutdown, or did this happen by itself? -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Sun Aug 31 14:51:32 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA20187 for smp-outgoing; Sun, 31 Aug 1997 14:51:32 -0700 (PDT) Received: from shell.futuresouth.com (shell.futuresouth.com [207.141.254.20]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA20182 for ; Sun, 31 Aug 1997 14:51:28 -0700 (PDT) Received: (from tim@localhost) by shell.futuresouth.com (8.8.5/8.8.5) id QAA04627; Sun, 31 Aug 1997 16:51:26 -0500 (CDT) Message-ID: <19970831165126.12973@futuresouth.com> Date: Sun, 31 Aug 1997 16:51:26 -0500 From: Tim Tsai To: freebsd-smp@freebsd.org Subject: fast clock with SMP kernel Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.76 Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk We have a TYAN TAHOE-II motherboard with a single Pentium-II-266 processor. When using 3.0-970807-SNAP with SMP options, the clock seems to run about twice as fast as it should. SMP-GENERIC seems to work fine. If I disable options SMP in our config file it works fine too. Is this because we only have one processor, or is this a different problem? The reason for running this kernel is that we're putting together a set of machines and some will have dual processors and I thought it's easier if we all run the same kernel. Thanks, Tim From owner-freebsd-smp Sun Aug 31 15:34:29 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA21862 for smp-outgoing; Sun, 31 Aug 1997 15:34:29 -0700 (PDT) Received: from news1.gtn.com (news1.gtn.com [192.109.159.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA21852 for ; Sun, 31 Aug 1997 15:34:16 -0700 (PDT) Received: (from uucp@localhost) by news1.gtn.com (8.8.6/8.8.6) with UUCP id AAA08694; Mon, 1 Sep 1997 00:15:19 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.7/8.8.7) id AAA03722; Mon, 1 Sep 1997 00:01:32 +0200 (CEST) Message-ID: <19970901000132.27494@klemm.gtn.com> Date: Mon, 1 Sep 1997 00:01:32 +0200 From: Andreas Klemm To: Steve Passe Cc: smp@FreeBSD.ORG Subject: Re: HEADS UP: another set of changes. References: <19970831231130.52301@klemm.gtn.com> <199708312125.PAA21626@Ilsa.StevesCafe.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79 In-Reply-To: <199708312125.PAA21626@Ilsa.StevesCafe.com>; from Steve Passe on Sun, Aug 31, 1997 at 03:25:58PM -0600 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT SMP Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sun, Aug 31, 1997 at 03:25:58PM -0600, Steve Passe wrote: > Hi, > > > Yesterday evening and today evening my system hangs with a page fault. > > > > I'm on CPU#1, I need to be on CPU#0, sleeping > > timeout waiting for CPU#0! > > syncing disks... > > Fatal trap 12: page fault while in kernel mode > > cpuid=1 > > lapic.id=0 > > fault virtual address = 0x10 > > fault code = supervisor-read, page not present > > instruction pointer = 0x8: 0xf0133110 > > stack pointer = 0x10:0xff803c5c > > frame pointer = 0x10:0xff803c70 > > code segment = base 0x0, limit 0xfffff, type=0x1b > > = DPL0, pres 1, def 32 1, gran 1 > > processor eflags = interrupt enabled, resume, IOPL = 0 > > current process = Idle > > mp_lock = 01000004 > > interrupt mask = tty <- SMP: XXX > > trap number = 12 > > panic: page fault > > cpuid 1 > > > > This was the output I noticed yestwerday evening. > > Today I cvsupped newly and build a new kernel. > > When the panic happened today at about 10pm, I compared the > > output with that above messages I wrote down, and everything > > is nearly the same, only the instruction stack and frame pointer > > were about off by one or two, because of the new kernel I think. > > > > What do you need to have more, so that you can figure out, what's > > going wrong ? > > we need a trace. also, it looks like it was shutting down when this happened? > specifically while syncing disks. had you down a shutdown, or did this happen > by itself? I wasn't at the machine ... X11 was running and processes like this: 11:59PM up 1:06, 4 users, load averages: 2.25, 2.22, 2.13 USER TTY FROM LOGIN@ IDLE WHAT andreas v0 - 10:57PM - xinit /home/andreas/.xinitr andreas p0 :0.0 10:57PM 8 -su (tcsh) andreas p2 :0.0 11:26PM 32 -tcsh (tcsh) andreas p3 :0.0 11:50PM - w USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME COMMAND root 240 52.0 0.3 388 156 con- RN 10:56PM 59:30.20 ./rc5v2 root 238 27.5 0.4 388 228 con- SN 10:56PM 61:03.27 ./rc5v2 root 1 0.0 0.1 480 44 ?? Is 12:53AM 0:00.03 /sbin/init -- root 2 0.0 0.0 0 0 ?? DL 12:53AM 0:12.60 (pagedaemon) root 3 0.0 0.0 0 0 ?? DL 12:53AM 0:00.00 (vmdaemon) root 4 0.0 0.0 0 0 ?? DL 12:53AM 0:02.32 (update) root 27 0.0 0.1 32488 68 ?? Is 12:56AM 0:00.21 mfs -o noatime -s 64000 /dev/sd1s1b /tmp (mount_mfs) root 33 0.0 0.0 204 0 ?? Is 12:56AM 0:00.00 adjkerntz -i root 72 0.0 0.5 224 288 ?? Ds 10:56PM 0:00.33 /usr/local/bin/bisdnd root 98 0.0 0.5 216 312 ?? Ss 10:56PM 0:00.29 syslogd root 104 0.0 0.5 408 288 ?? Ss 10:56PM 0:00.04 named -b /etc/namedb/named.boot root 127 0.0 0.3 192 176 ?? Is 10:56PM 0:00.05 inetd root 129 0.0 0.4 332 228 ?? Is 10:56PM 0:00.10 cron root 132 0.0 0.2 212 112 ?? Is 10:56PM 0:00.01 lpd root 135 0.0 0.4 564 252 ?? Is 10:56PM 0:00.06 sendmail: accepting connections on port 25 (sendmail) root 193 0.0 0.4 912 252 ?? Ss 10:56PM 0:00.50 /usr/local/sbin/httpd root 206 0.0 0.2 588 92 con- I 10:56PM 0:00.03 /usr/local/sbin/rplayd root 211 0.0 0.0 492 0 con- I 10:56PM 0:00.01 /bin/sh /usr/local/sbin/RunCache nobody 218 0.0 1.2 1928 776 con- S 10:56PM 0:01.75 squid -sY root 219 0.0 0.3 316 156 ?? Is 10:56PM 0:00.84 /usr/local/sbin/sshd nobody 229 0.0 0.2 264 108 con- I 10:56PM 0:00.02 (dnsserver) (dnsserver) nobody 237 0.0 0.2 236 104 ?? Is 10:56PM 0:00.01 /usr/local/sbin/ftpget -S 1026 nobody 246 0.0 0.1 136 52 ?? I 10:56PM 0:00.01 (unlinkd) (unlinkd) andreas 255 0.0 0.2 808 128 v0 Is 10:56PM 0:00.33 -tcsh (tcsh) root 256 0.0 0.2 180 96 v1 Is+ 10:56PM 0:00.02 /usr/libexec/getty Pc ttyv1 root 257 0.0 0.2 180 96 v2 Is+ 10:56PM 0:00.02 /usr/libexec/getty Pc ttyv2 root 258 0.0 0.2 180 96 v3 Is+ 10:56PM 0:00.02 /usr/libexec/getty Pc ttyv3 root 261 0.0 0.2 196 116 ?? S 10:56PM 0:00.06 (pinger) (pinger) news 279 0.0 1.3 1580 844 ?? Is 10:56PM 0:01.90 /usr/local//etc/innd -p4 -r -i0 root 280 0.0 0.0 496 0 ?? I 10:56PM 0:00.00 /bin/sh /usr/local/etc/rc.news root 282 0.0 0.2 512 112 ?? I 10:56PM 0:03.20 /bin/sh /usr/local/news/bin/innwatch news 285 0.0 0.2 152 92 ?? I 10:56PM 0:00.09 /usr/local/news/bin/overchan andreas 292 0.0 0.3 176 152 v0 I+ 10:57PM 0:00.08 xinit /home/andreas/.xinitrc -- -auth /home/andreas/.Xauthority -bpp 16 root 305 0.0 10.6 5224 6684 ?? S 10:57PM 0:14.15 X :0 -auth /home/andreas/.Xauthority -bpp 16 (XF86_SVGA) andreas 308 0.0 0.0 500 0 v0 I 10:57PM 0:00.03 sh /home/andreas/.xinitrc andreas 318 0.0 0.6 424 388 v0 I 10:57PM 0:00.70 /usr/X11R6/bin/afterstep andreas 319 0.0 0.3 180 192 v0 I 10:57PM 0:00.04 /usr/X11R6/lib/X11/afterstep/Animate 8 5 /home/andreas/.steprc 0 0 andreas 320 0.0 0.4 280 252 v0 I 10:57PM 0:00.28 /usr/X11R6/lib/X11/afterstep/Wharf 10 5 /home/andreas/.steprc 0 0 andreas 321 0.0 0.2 156 108 v0 I 10:57PM 0:00.04 /usr/X11R6/lib/X11/afterstep/Audio 12 5 /home/andreas/.steprc 0 0 andreas 323 0.0 0.4 204 244 v0 S 10:57PM 0:05.98 asclock -exe xlock -12 -shape -position -1-1 andreas 325 0.0 0.6 212 340 v0 S 10:57PM 0:00.15 xbiff andreas 327 0.0 0.6 252 352 v0 S 10:57PM 0:00.25 xload andreas 329 0.0 0.5 208 336 v0 I 10:57PM 0:00.12 /usr/X11R6/lib/X11/afterstep/Pager 13 4 /home/andreas/.steprc 0 8 0 0 andreas 330 0.0 0.8 292 472 v0 I 10:57PM 0:00.17 xpostit -T xpostit root 337 0.0 1.0 744 656 v0 I 10:57PM 0:00.71 xterm -e su - root 339 0.0 0.6 1084 388 p0 Is+ 10:57PM 0:00.60 -su (tcsh) root 381 0.0 1.2 256 744 p0 S 10:57PM 0:00.23 xconsole root 1631 0.0 0.5 344 300 ?? Ss 11:00PM 0:00.20 nmbd nobody 1656 0.0 0.5 976 308 ?? I 11:03PM 0:00.06 /usr/local/sbin/httpd root 2399 0.0 1.0 744 600 v0 I 11:26PM 0:00.16 xterm andreas 2400 0.0 0.3 808 164 p2 Is+ 11:26PM 0:00.24 -tcsh (tcsh) andreas 2408 0.0 4.6 14804 2912 v0 I 11:26PM 0:04.01 /usr/local/lib/netscape/netscape.bin andreas 2410 0.0 0.6 10844 360 v0 I 11:26PM 0:00.09 (dns helper) (netscape.bin) nobody 2415 0.0 0.5 972 308 ?? I 11:27PM 0:00.02 /usr/local/sbin/httpd nobody 2418 0.0 0.4 912 252 ?? I 11:27PM 0:00.00 /usr/local/sbin/httpd root 3388 0.0 2.6 508 1624 v0 S 11:50PM 0:00.35 xterm -e mutt andreas 3389 0.0 1.4 700 852 p3 Is+ 11:50PM 0:00.23 mutt root 3671 0.0 0.0 172 12 ?? I 11:58PM 0:00.00 sleep 600 andreas 3672 0.0 0.3 496 172 p3 I+ 11:58PM 0:00.01 sh -c /usr/bin/vi '/tmp/mutt-titan-3389-2' andreas 3673 0.0 1.2 676 736 p3 S+ 11:58PM 0:00.17 /usr/bin/vi /tmp/mutt-titan-3389-2 andreas 3679 0.0 1.4 744 900 p3 S+ 11:59PM 0:00.17 tcsh -c ps -axuww andreas 3680 0.0 0.0 0 0 p3 Z+ - 0:00.00 (vi) andreas 3683 0.0 0.5 672 304 p3 R+ 11:59PM 0:00.00 ps -axuww root 0 0.0 0.0 0 0 ?? DLs 12:53AM 0:00.01 (swapper) When I turned on the monitor I only saw the panic after it happened .... So what now, do I need a serial console, because I always run X11 or do I have to simply compile DDB into the kernel ??? Ot additionally config -g ??? Andreas /// -- Andreas Klemm | klemm.gtn.com - powered by Symmetric MultiProcessor FreeBSD http://www.freebsd.org/~fsmp/SMP/SMP.html http://www.freebsd.org/~fsmp/SMP/benches.html From owner-freebsd-smp Sun Aug 31 15:34:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA21840 for smp-outgoing; Sun, 31 Aug 1997 15:34:12 -0700 (PDT) Received: from news1.gtn.com (news1.gtn.com [192.109.159.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA21794 for ; Sun, 31 Aug 1997 15:34:04 -0700 (PDT) Received: (from uucp@localhost) by news1.gtn.com (8.8.6/8.8.6) with UUCP id AAA14155; Mon, 1 Sep 1997 00:30:17 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.7/8.8.7) id AAA00615; Mon, 1 Sep 1997 00:17:17 +0200 (CEST) Message-ID: <19970901001717.08535@klemm.gtn.com> Date: Mon, 1 Sep 1997 00:17:17 +0200 From: Andreas Klemm To: Steve Passe Cc: smp@FreeBSD.ORG Subject: Re: HEADS UP: another set of changes. References: <19970831231130.52301@klemm.gtn.com> <199708312125.PAA21626@Ilsa.StevesCafe.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79 In-Reply-To: <199708312125.PAA21626@Ilsa.StevesCafe.com>; from Steve Passe on Sun, Aug 31, 1997 at 03:25:58PM -0600 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT SMP Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sun, Aug 31, 1997 at 03:25:58PM -0600, Steve Passe wrote: > > we need a trace. also, it looks like it was shutting down when this happened? > specifically while syncing disks. had you down a shutdown, or did this happen > by itself? I compiled a new kernel using DDB as The kernel is loaded and after that the screen gets dark and the machine performs a total reset .. -- Andreas Klemm | klemm.gtn.com - powered by Symmetric MultiProcessor FreeBSD http://www.freebsd.org/~fsmp/SMP/SMP.html http://www.freebsd.org/~fsmp/SMP/benches.html From owner-freebsd-smp Sun Aug 31 15:40:07 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA22153 for smp-outgoing; Sun, 31 Aug 1997 15:40:07 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA22147 for ; Sun, 31 Aug 1997 15:40:05 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id QAA22012; Sun, 31 Aug 1997 16:39:54 -0600 (MDT) Message-Id: <199708312239.QAA22012@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Andreas Klemm cc: smp@FreeBSD.ORG Subject: Re: HEADS UP: another set of changes. In-reply-to: Your message of "Mon, 01 Sep 1997 00:17:17 +0200." <19970901001717.08535@klemm.gtn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 31 Aug 1997 16:39:53 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk try using an unmodified (as possible) SMP-GENERIC config file. any lkms being loaded? they seem to be causing trouble again... -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Sun Aug 31 19:56:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id TAA01688 for smp-outgoing; Sun, 31 Aug 1997 19:56:13 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id TAA01679 for ; Sun, 31 Aug 1997 19:56:10 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id UAA22893; Sun, 31 Aug 1997 20:56:07 -0600 (MDT) Message-Id: <199709010256.UAA22893@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Rick Lotoczky cc: smp@FreeBSD.ORG Subject: Re: SMP & P6DOF - Show of Hands In-reply-to: Your message of "Sun, 31 Aug 1997 16:09:33 EDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 31 Aug 1997 20:56:07 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > Sorry for the late reply, but I was out of town for the last week. Seems > like the changes you made to SMP system were a big win here. Silo overflows > are non-existant, a few interrupt-overflows remain. I can now access the > IDE drives **and** connect with my 2B ISDN line concurrently. In the past, > the sio would all but stop during a disk transfer. to eliminate tty-level buffer overflows patch sys/sys/tty.h: *** tty.h.orig 1997/08/31 23:03:59 1.1 --- tty.h 1997/08/31 23:04:52 *************** *** 119,125 **** --- 119,129 ---- #define OBUFSIZ 100 #ifndef TTYHOG + #ifdef SMP + #define TTYHOG 2048 + #else #define TTYHOG 1024 + #endif /* SMP */ #endif #ifdef KERNEL Send reports of success/failure. If it does some good I'll ccommit it. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Sun Aug 31 23:04:21 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA08755 for smp-outgoing; Sun, 31 Aug 1997 23:04:21 -0700 (PDT) Received: from news1.gtn.com (news1.gtn.com [192.109.159.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA08750 for ; Sun, 31 Aug 1997 23:04:12 -0700 (PDT) Received: (from uucp@localhost) by news1.gtn.com (8.8.6/8.8.6) with UUCP id IAA11395; Mon, 1 Sep 1997 08:00:15 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.7/8.8.7) id HAA00850; Mon, 1 Sep 1997 07:45:41 +0200 (CEST) Message-ID: <19970901074541.33637@klemm.gtn.com> Date: Mon, 1 Sep 1997 07:45:41 +0200 From: Andreas Klemm To: Steve Passe Cc: smp@FreeBSD.ORG Subject: Re: HEADS UP: another set of changes. References: <19970901001717.08535@klemm.gtn.com> <199708312239.QAA22012@Ilsa.StevesCafe.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79 In-Reply-To: <199708312239.QAA22012@Ilsa.StevesCafe.com>; from Steve Passe on Sun, Aug 31, 1997 at 04:39:53PM -0600 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT SMP Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sun, Aug 31, 1997 at 04:39:53PM -0600, Steve Passe wrote: > try using an unmodified (as possible) SMP-GENERIC config file. Ok ... > any lkms being loaded? they seem to be causing trouble again... Yes Type Id Off Loadaddr Size Info Rev Module Name EXEC 0 4 f57d8000 001c f57de010 1 linux_mod MISC 1 0 f57df000 0008 f57e0000 1 daemon_saver_mod And here my kernel config file ... # # BISDN kernel # machine "i386" cpu "I686_CPU" ident BISDNSMP maxusers 64 # SMP Stuff options SMP # Symmetric MultiProcessor Kernel options APIC_IO # Symmetric (APIC) I/O options USER_LDT # for Wine options SHOW_BUSYBUFS # busy buffers on shutdown ? # Networking options INET #InterNETworking options IPFIREWALL #firewall options IPFIREWALL_VERBOSE #print information about dropped packets options "IPFIREWALL_VERBOSE_LIMIT=100" #limit verbosity # filesystems options FFS #Berkeley Fast Filesystem options MFS #Memory File System options PROCFS #Process filesystem options MSDOSFS #MS DOS File System options "CD9660" #ISO 9660 filesystem options NSWAPDEV=3 #Allow this many swap-devices. # misc options options "COMPAT_43" #Compatible with BSD 4.3 [KEEP THIS!] options UCONSOLE #Allow users to grab the console options SYSVSHM,SYSVSEM,SYSVMSG #shared memory (X11) options "MD5" options COMPAT_LINUX # Linux Binary compatibility config kernel root on sd1 # ISA and PCI BUS support controller isa0 controller pci0 # Floppy Disk Controller controller fdc0 at isa? port "IO_FD1" bio irq 6 drq 2 vector fdintr disk fd0 at fdc0 drive 0 # AHA 2940 PCI Controller controller ahc0 # SCSI Devices controller scbus0 device sd0 # Harddisk 0 - DOS/FreeBSD SMP device sd1 # Harddisk 1 - FreeBSD Boot device sd2 # Harddisk 2 - FreeBSD local device st0 # TDC 4222 device cd0 # TOSHIBA XM-5701TA 3136 options AHC_TAGENABLE # tagged command queueing options AHC_ALLOW_MEMIO options AHC_SCBPAGING_ENABLE options SCSI_REPORT_GEOMETRY # SCO compatible system console device sc0 at isa? port "IO_KBD" tty irq 1 vector scintr options MAXCONS=4 # number of virtual consoles # floating point unit device npx0 at isa? port "IO_NPX" flags 0x1 irq 13 vector npxintr # serial devices on mainboard device sio0 at isa? port "IO_COM1" tty irq 4 vector siointr device sio1 at isa? port "IO_COM2" tty irq 3 vector siointr # parallel device on mainboard device lpt0 at isa? port? tty irq 7 vector lptintr # PS/2 mouse on mainboard device psm0 at isa? port "IO_KBD" conflicts tty irq 12 vector psmintr options "PSM_ACCEL=1" # PS/2 mouse acceleration # Joystick device joy0 at isa? port "IO_GAME" # Network 3COM PCI # device vx0 device ed0 at isa? port 0x280 net irq 10 iomem 0xd0000 vector edintr # Soundblaster 16 # SoundBlaster DSP driver - for SB, SB Pro, SB16, PAS(emulating SB) # SoundBlaster 16 DSP driver - for SB16 - requires sb0 device # SoundBlaster 16 MIDI - for SB16 - requires sb0 device # Yamaha OPL-2/OPL-3 FM - for SB, SB Pro, SB16, PAS controller snd0 device sb0 at isa? port 0x220 irq 5 conflicts drq 1 vector sbintr device sbxvi0 at isa? port? irq? drq 5 conflicts device sbmidi0 at isa? port 0x330 irq? conflicts device opl0 at isa? port 0x388 conflicts # Pseudo devices pseudo-device loop pseudo-device ether pseudo-device log #Kernel syslog interface (/dev/klog) pseudo-device vn 1 #Vnode driver (turns a file into a dev.) pseudo-device tun 1 #user mode ppp pseudo-device bpfilter 2 #Berkeley packet filter pseudo-device pty 16 pseudo-device gzip # Exec gzipped a.out's # BISDN options IPI_VJ # Van Jacobsen header compression support #options "IPI_DIPA=3" # send ip accounting packets every 3 seconds options TELES_HAS_MEMCPYB # bisdn 0.97 # Teles S0/16.3 ###################################################### IRQ 9 ## controller tel0 at isa? port 0xd80 net irq 9 vector telintr pseudo-device disdn pseudo-device isdn pseudo-device ipi 1 pseudo-device ispy 1 -- Andreas Klemm | klemm.gtn.com - powered by Symmetric MultiProcessor FreeBSD http://www.freebsd.org/~fsmp/SMP/SMP.html http://www.freebsd.org/~fsmp/SMP/benches.html From owner-freebsd-smp Mon Sep 1 04:41:36 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id EAA21389 for smp-outgoing; Mon, 1 Sep 1997 04:41:36 -0700 (PDT) Received: from ic.net (qmailr@srv1b.ic.net [152.160.72.20]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id EAA21382 for ; Mon, 1 Sep 1997 04:41:34 -0700 (PDT) Received: (qmail 20534 invoked from network); 1 Sep 1997 11:23:19 -0000 Received: from unknown (HELO lurch.rickl.org) (152.160.108.46) by unknown with SMTP; 1 Sep 1997 11:23:19 -0000 Received: from lurch.rickl.org (rickl@lurch.rickl.org [192.168.255.1]) by lurch.rickl.org (8.8.7/8.7.3) with ESMTP id HAA00290; Mon, 1 Sep 1997 07:22:32 -0400 (EDT) Date: Mon, 1 Sep 1997 07:22:32 -0400 (EDT) From: Rick Lotoczky Reply-To: Rick Lotoczky Subject: sio overflows To: smp@csn.net cc: smp@freebsd.org Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi Steve, I tried the patch you sent and unfortunately it seemed to have little effect. I even tried increasing the size of TTYHOG to 4096 and had essentially the same results. To give a little more insight, the overflows occur during disk access and when I swap virtual screens in X and FVWM. My system has 2 IDE drives, used mainly for the system (wd0) and archive storage (wd1). Also, /tmp is on wd0. I have 3 SCSI drives, 1 for swap, and the other two as working drives (/usr) and (/home). The /home drive is my home directory with all the application temp files, etc. on it. The serial ports are 16650 types (I have flags=0x200000 set on the sio config line) with a 230K link to a Bitsurfer Pro. Thanks for the help Rick From owner-freebsd-smp Mon Sep 1 04:52:15 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id EAA21929 for smp-outgoing; Mon, 1 Sep 1997 04:52:15 -0700 (PDT) Received: from silvia.myaw.ei.meisei-u.ac.jp (silvia.myaw.ei.meisei-u.ac.jp [160.194.103.5]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id EAA21916 for ; Mon, 1 Sep 1997 04:52:09 -0700 (PDT) Received: from silvia.myaw.ei.meisei-u.ac.jp (localhost [127.0.0.1]) by silvia.myaw.ei.meisei-u.ac.jp (8.8.6/8.8.5) with ESMTP id UAA10588 for ; Mon, 1 Sep 1997 20:51:55 +0900 (JST) Message-Id: <199709011151.UAA10588@silvia.myaw.ei.meisei-u.ac.jp> To: smp@freebsd.org Subject: Fatal trap 18: integer divide fault while in kernel mode From: Akira Watanabe X-Mailer: Mew version 1.70 on Emacs 19.28.1 / Mule 2.3 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Date: Mon, 01 Sep 1997 20:51:55 +0900 Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi. The kernel (suped yesterday) causes a panic. Fatal trap 18: integer divide fault while in kernel mode cpuid = 0 lapic.id = 16777216 instruction pointer = 0x8:0xf01bc794 stack pointer = 0x10:0xf4cabc84 frame pointer = 0x10:0xf4cabcd0 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 236 (ftpd) mp_lock = 00000003 interrupt mask = <- SMP: XXX trap number = 18 panic: integer divide fault cpuid 0 boot() called on cpu#0 syncing disks... 11 11 8 2 done Here is a stack trace. # gdb -k kernel /var/crash/vmcore.0 GDB is free software and you are welcome to distribute copies of it under certain conditions; type "show copying" to see the conditions. There is absolutely no warranty for GDB; type "show warranty" for details. GDB 4.16 (i386-unknown-freebsd), Copyright 1996 Free Software Foundation, Inc... IdlePTD 24d000 current pcb at 1f9608 panic: integer divide fault #0 boot (howto=256) at ../../kern/kern_shutdown.c:289 289 dumppcb.pcb_cr3 = rcr3(); (kgdb) where #0 boot (howto=256) at ../../kern/kern_shutdown.c:289 #1 0xf0118e36 in panic (fmt=0xf01ccbea "integer divide fault") at ../../kern/kern_shutdown.c:416 #2 0xf01cd86f in trap_fatal (frame=0xf4cabc48) at ../../i386/i386/trap.c:806 #3 0xf01cd072 in trap (frame={tf_es = -256049136, tf_ds = 131088, tf_edi = -256677376, tf_esi = 0, tf_ebp = -188039984, tf_isp = -188040080, tf_ebx = 0, tf_edx = 0, tf_ecx = 4096, tf_eax = 4096, tf_trapno = 18, tf_err = 0, tf_eip = -266614892, tf_cs = 8, tf_eflags = 66118, tf_esp = 0, tf_ss = 3}) at ../../i386/i386/trap.c:487 #4 0xf01bc794 in vnode_pager_haspage (object=0xf0bdb800, pindex=0, before=0xf4cabd34, after=0xf4cabd30) at ../../vm/vnode_pager.c:231 #5 0xf01bbcff in vm_pager_has_page (object=0xf0bdb800, offset=0, before=0xf4cabd34, after=0xf4cabd30) at ../../vm/vm_pager.c:205 #6 0xf01b2e05 in vm_fault_additional_pages (m=0xf04ef7c4, rbehind=3, rahead=4, marray=0xf4cabdd0, reqpage=0xf4cabda4) at ../../vm/vm_fault.c:1100 #7 0xf01b21c0 in vm_fault (map=0xf0bd9300, vaddr=134385664, fault_type=1 '\001', fault_flags=0) at ../../vm/vm_fault.c:414 #8 0xf01cd23a in trap_pfault (frame=0xf4cabe50, usermode=0) at ../../i386/i386/trap.c:681 #9 0xf01ccf47 in trap (frame={tf_es = 134348816, tf_ds = 134348816, tf_edi = -259133440, tf_esi = 134385664, tf_ebp = -188039496, tf_isp = -188039560, tf_ebx = 2048, tf_edx = 134387712, tf_ecx = 512, tf_eax = -188047360, tf_trapno = 12, tf_err = 0, tf_eip = -266551371, tf_cs = 8, tf_eflags = 66054, tf_esp = -188039368, tf_ss = -188039376}) at ../../i386/i386/trap.c:339 #10 0xf01cbfb5 in generic_copyin () #11 0xf012cf6f in sosend (so=0xf0bddc00, addr=0x0, uio=0xf4cabf38, top=0x0, control=0x0, flags=0, p=0xf0bb9600) at ../../kern/uipc_socket.c:449 #12 0xf0122ec8 in soo_write (fp=0xf0bdcd40, uio=0xf4cabf38, cred=0xf0bd8b00) at ../../kern/sys_socket.c:78 #13 0xf0120884 in write (p=0xf0bb9600, uap=0xf4cabf94, retval=0xf4cabf84) at ../../kern/sys_generic.c:268 #14 0xf01cdacb in syscall (frame={tf_es = 39, tf_ds = 39, tf_edi = 134385664, tf_esi = 2256, tf_ebp = -272642012, tf_isp = -188039196, tf_ebx = 6, tf_edx = 5, tf_ecx = 1, tf_eax = 4, tf_trapno = 22, tf_err = 7, tf_eip = 135028641, tf_cs = 31, tf_eflags = 531, tf_esp = -272642064, tf_ss = 39}) at ../../i386/i386/trap.c:953 #15 0x80c5fa1 in ?? () #16 0x3658 in ?? () #17 0x86fb in ?? () #18 0x2045 in ?? () #19 0x1096 in ?? () (kgdb) Another problem is that sometimes the kernel reports like de0: abnormal interrupt: transmit underflow (raising TX threshold to 8|512) Is this related to a change of the locking scheme under SMP ? ------- Akira Watanabe, Meisei University. From owner-freebsd-smp Mon Sep 1 09:31:59 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA06686 for smp-outgoing; Mon, 1 Sep 1997 09:31:59 -0700 (PDT) Received: from tasogare.imasy.or.jp (root@tasogare.imasy.or.jp [202.227.24.5]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA06680 for ; Mon, 1 Sep 1997 09:31:55 -0700 (PDT) Received: (from ume@localhost) by tasogare.imasy.or.jp (8.8.7+2.7Wbeta7/3.4W4-96030215) with UUCP id BAA10070; Tue, 2 Sep 1997 01:27:06 +0900 (JST) Received: from peace.calm.imasy.or.jp (root@peace.calm.imasy.or.jp [158.214.107.233]) by chaos.calm.imasy.or.jp (8.8.7/3.6Wbeta6-CHAOS1.5) with ESMTP id BAA05625; Tue, 2 Sep 1997 01:24:26 +0900 (JST) Received: from localhost (ume@localhost [127.0.0.1]) by peace.calm.imasy.or.jp (8.8.7/3.6Wbeta6-CALM1.0) with ESMTP id BAA09997; Tue, 2 Sep 1997 01:23:55 +0900 (JST) Message-Id: <199709011623.BAA09997@peace.calm.imasy.or.jp> To: smp@csn.net Cc: smp@neutron.neutron.org, FreeBSD-SMP@FreeBSD.ORG Subject: Re: motherboard choice In-Reply-To: Your message of "Sun, 31 Aug 1997 15:14:41 -0600" <199708312114.PAA21499@Ilsa.StevesCafe.com> References: <199708312114.PAA21499@Ilsa.StevesCafe.com> X-Mailer: Mew version 1.90 on XEmacs 20.3 (Bratislava) X-PGP-Fingerprint: 6B 0C 53 FC 5D D0 37 91 05 D0 B3 EF 36 9B 6A BC X-URL: http://www.imasy.or.jp/~ume/ Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Date: Tue, 02 Sep 1997 01:23:55 +0900 From: Hajimu UMEMOTO X-Dispatcher: imput version 970830 Lines: 15 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, >>>>> On Sun, 31 Aug 1997 15:14:41 -0600, Steve Passe said: smp> I personally suggest the gigabyte ga686dx over the tyan. Iwill makes smp> a dual P6 with the adaptec UW on it, but I'm not sure if anyone has reported smp> success with it yet. I'm using IWILL DP6NS w/ P6-180 x 2. It seems to work fine for me. -- Hajimu UMEMOTO @ Internet Mutual Aid Society Yokohama, Japan ume@imasy.or.jp ume@iabs.hitachi.co.jp http://www.imasy.or.jp/~ume/ From owner-freebsd-smp Mon Sep 1 10:20:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id KAA10186 for smp-outgoing; Mon, 1 Sep 1997 10:20:13 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id KAA10173 for ; Mon, 1 Sep 1997 10:20:09 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id LAA25798; Mon, 1 Sep 1997 11:20:06 -0600 (MDT) Message-Id: <199709011720.LAA25798@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Rick Lotoczky cc: smp@freebsd.org Subject: Re: sio overflows In-reply-to: Your message of "Mon, 01 Sep 1997 07:22:32 EDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 01 Sep 1997 11:20:06 -0600 Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk HI, > I tried the patch you sent and unfortunately it seemed to have little > effect. I even tried increasing the size of TTYHOG to 4096 and had > essentially the same results. To give a little more insight, the overflows > occur during disk access and when I swap virtual screens in X and FVWM. > My system has 2 IDE drives, used mainly for the system (wd0) and archive > storage (wd1). Also, /tmp is on wd0. I have 3 SCSI drives, 1 for swap, > and the other two as working drives (/usr) and (/home). The /home drive > is my home directory with all the application temp files, etc. on it. The > serial ports are 16650 types (I have flags=0x200000 set on the sio config > line) with a 230K link to a Bitsurfer Pro. do you use hardware flow control? do you have any problem with same setup under UP kernel? try undefining srs/sys/i386/include/smptests.h: FAST_WITHOUTCPL -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Mon Sep 1 14:56:20 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA21928 for smp-outgoing; Mon, 1 Sep 1997 14:56:20 -0700 (PDT) Received: from ic.net (qmailr@srv2b.ic.net [152.160.72.21]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id OAA21917 for ; Mon, 1 Sep 1997 14:56:14 -0700 (PDT) Received: (qmail 10097 invoked from network); 1 Sep 1997 21:35:45 -0000 Received: from unknown (HELO lurch.rickl.org) (152.160.108.49) by unknown with SMTP; 1 Sep 1997 21:35:45 -0000 Received: from lurch.rickl.org (rickl@lurch.rickl.org [192.168.255.1]) by lurch.rickl.org (8.8.7/8.7.3) with ESMTP id RAA04955; Mon, 1 Sep 1997 17:35:12 -0400 (EDT) Date: Mon, 1 Sep 1997 17:35:11 -0400 (EDT) From: Rick Lotoczky Reply-To: Rick Lotoczky Subject: Re: sio overflows To: smp@csn.net cc: smp@FreeBSD.ORG In-Reply-To: <199709011720.LAA25798@Ilsa.StevesCafe.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk HI, >>My stuff deleted >do you use hardware flow control? Yup, sure do . >do you have any problem with same setup under UP kernel? Didn't in the past, but I haven't tried it since the SMP version is working so well. >try undefining srs/sys/i386/include/smptests.h: >FAST_WITHOUTCPL OK, will try it tonite and let you know what happens. Thanks again -- Rick From owner-freebsd-smp Mon Sep 1 22:25:09 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA15433 for smp-outgoing; Mon, 1 Sep 1997 22:25:09 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA15422 for ; Mon, 1 Sep 1997 22:25:03 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id XAA27888 for ; Mon, 1 Sep 1997 23:25:02 -0600 (MDT) Message-Id: <199709020525.XAA27888@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: smp@freebsd.org Subject: HEADS UP: good time to update. Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 01 Sep 1997 23:25:01 -0600 Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, I am close to finishing the removal of the giant lock on regular (ie non-FAST) INTerrupts. This is likely to cause some new bugs during its shakeout period. Recent reports are that the current SMP code is fairly stable (except for that Fatal trap 12 problem). So it might be a good time to update. In particular the current code seems to be a big win for sio and IDE. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Mon Sep 1 22:59:45 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA17160 for smp-outgoing; Mon, 1 Sep 1997 22:59:45 -0700 (PDT) Received: from monk.via.net (monk.via.net [140.174.204.10]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id WAA17154 for ; Mon, 1 Sep 1997 22:59:43 -0700 (PDT) Received: (from joe@localhost) by monk.via.net (8.6.11/8.6.12) id WAA15138 for smp@FreeBSD.ORG; Mon, 1 Sep 1997 22:50:05 -0700 Date: Mon, 1 Sep 1997 22:50:05 -0700 From: Joe McGuckin Message-Id: <199709020550.WAA15138@monk.via.net> To: smp@FreeBSD.ORG Subject: Multi threaded FreeBSD kernel? X-Sun-Charset: US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Will it ever happen? From owner-freebsd-smp Tue Sep 2 01:56:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id BAA25979 for smp-outgoing; Tue, 2 Sep 1997 01:56:04 -0700 (PDT) Received: from firewall.ftf.dk (root@mail.ftf.dk [129.142.64.2]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id BAA25974 for ; Tue, 2 Sep 1997 01:56:00 -0700 (PDT) Received: from mail.prosa.dk ([192.168.100.2]) by firewall.ftf.dk (8.7.6/8.7.3) with ESMTP id LAA04514; Tue, 2 Sep 1997 11:25:01 +0200 Received: from deepo.prosa.dk (deepo.prosa.dk [192.168.100.10]) by mail.prosa.dk (8.8.5/8.8.5/prosa-1.1) with ESMTP id LAA20811; Tue, 2 Sep 1997 11:00:44 +0200 (CEST) Received: (from regnauld@localhost) by deepo.prosa.dk (8.8.5/8.8.5/prosa-1.1) id KAA07686; Tue, 2 Sep 1997 10:54:27 +0200 (CEST) Message-ID: <19970902105427.12627@deepo.prosa.dk> Date: Tue, 2 Sep 1997 10:54:27 +0200 From: Philippe Regnauld To: Joe McGuckin Cc: smp@FreeBSD.ORG Subject: Re: Multi threaded FreeBSD kernel? References: <199709020550.WAA15138@monk.via.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Description: Main Body X-Mailer: Mutt 0.69 In-Reply-To: <199709020550.WAA15138@monk.via.net>; from Joe McGuckin on Mon, Sep 01, 1997 at 10:50:05PM -0700 X-Operating-System: FreeBSD 2.2.1-RELEASE i386 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Joe McGuckin writes: > > Will it ever happen? Are you working on it ? :-) Like Phk said recently (quite close, but not sure :-) "There are about 10% of programmers who program in something else than Cobol, Delphi or Visual Basic. Of those 10%, some work in C, and of that percentage about 10% do kernel development. Of those 10%, 10% are capable of programming and understanding threads, and 1% of those like it". -- -- Phil -[ Philippe Regnauld / Systems Administrator / regnauld@deepo.prosa.dk ]- -[ Location.: +55.4N +11.3E PGP Key: finger regnauld@hotel.prosa.dk ]- From owner-freebsd-smp Tue Sep 2 02:36:36 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id CAA27531 for smp-outgoing; Tue, 2 Sep 1997 02:36:36 -0700 (PDT) Received: from ic.net (qmailr@srv1b.ic.net [152.160.72.20]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id CAA27526 for ; Tue, 2 Sep 1997 02:36:31 -0700 (PDT) Received: (qmail 20571 invoked from network); 2 Sep 1997 09:36:19 -0000 Received: from unknown (HELO lurch.rickl.org) (152.160.108.8) by unknown with SMTP; 2 Sep 1997 09:36:19 -0000 Received: from lurch.rickl.org (rickl@lurch.rickl.org [192.168.255.1]) by lurch.rickl.org (8.8.7/8.7.3) with ESMTP id FAA00831; Tue, 2 Sep 1997 05:35:43 -0400 (EDT) Date: Tue, 2 Sep 1997 05:35:43 -0400 (EDT) From: Rick Lotoczky Reply-To: Rick Lotoczky Subject: sio stuff To: smp@csn.net cc: smp@freebsd.org Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi Steve, I tried to undef FAST_WITHOUTCPL in sys/i386/include/smptests.h and it hangs the system immediately when trying to invoke a serial comm link and immediately when invoking X. I did not get any debugging output since I only have a single local monitor. If there is anything else you would like me to try, let me know. Rick From owner-freebsd-smp Tue Sep 2 11:21:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA19611 for smp-outgoing; Tue, 2 Sep 1997 11:21:12 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA19603 for ; Tue, 2 Sep 1997 11:21:08 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id MAA00339; Tue, 2 Sep 1997 12:20:47 -0600 (MDT) Message-Id: <199709021820.MAA00339@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Tim Tsai cc: freebsd-smp@FreeBSD.ORG Subject: Re: fast clock with SMP kernel In-reply-to: Your message of "Sun, 31 Aug 1997 16:51:26 CDT." <19970831165126.12973@futuresouth.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Sep 1997 12:20:47 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > We have a TYAN TAHOE-II motherboard with a single Pentium-II-266 processor. > When using 3.0-970807-SNAP with SMP options, the clock seems to run about > twice as fast as it should. SMP-GENERIC seems to work fine. If I disable > options SMP in our config file it works fine too. Is this because we only > have one processor, or is this a different problem? I'm surprised it runs at all on one CPU! If you disable options SMP you get a UP kernel, so thats not relevant. The real question is why the difference between 3.0-970807-SNAP with SMP options and SMP-GENERIC. I have no answer. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Tue Sep 2 11:22:42 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA19699 for smp-outgoing; Tue, 2 Sep 1997 11:22:42 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA19692 for ; Tue, 2 Sep 1997 11:22:38 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id MAA00351; Tue, 2 Sep 1997 12:22:28 -0600 (MDT) Message-Id: <199709021822.MAA00351@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Akira Watanabe cc: smp@FreeBSD.ORG Subject: Re: Fatal trap 18: integer divide fault while in kernel mode In-reply-to: Your message of "Mon, 01 Sep 1997 20:51:55 +0900." <199709011151.UAA10588@silvia.myaw.ei.meisei-u.ac.jp> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Sep 1997 12:22:28 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, show of hands please, who is still getting "fatal trap 12" errors? -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Tue Sep 2 11:26:41 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA19956 for smp-outgoing; Tue, 2 Sep 1997 11:26:41 -0700 (PDT) Received: from wile-e-coyote.cs.washington.edu (wile-e-coyote.cs.washington.edu [128.95.2.83]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA19949 for ; Tue, 2 Sep 1997 11:26:38 -0700 (PDT) From: mef@cs.washington.edu Received: (mef@localhost) by wile-e-coyote.cs.washington.edu (8.8.5+CS/7.2ws+) id LAA13628; Tue, 2 Sep 1997 11:26:46 -0700 (PDT) Date: Tue, 2 Sep 1997 11:26:46 -0700 (PDT) Message-Id: <199709021826.LAA13628@wile-e-coyote.cs.washington.edu> To: smp@freebsd.org Subject: smp motherboard w/ i960/i2o support Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, Suppose you could order any smp system (motherboard w/ an i960 processor) on it, and $$$ wasn't really an issue, then which one would you choose? From what I can gather fbsd-smp has some problems with the Supermicro P6DNH. What other smp boards w/i960 are out there and are good ones to use with fbsd-smp? Thanks, Marc From owner-freebsd-smp Tue Sep 2 12:13:21 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id MAA21913 for smp-outgoing; Tue, 2 Sep 1997 12:13:21 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id MAA21905 for ; Tue, 2 Sep 1997 12:13:18 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id NAA00552; Tue, 2 Sep 1997 13:13:07 -0600 (MDT) Message-Id: <199709021913.NAA00552@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: mef@cs.washington.edu cc: smp@FreeBSD.ORG Subject: Re: smp motherboard w/ i960/i2o support In-reply-to: Your message of "Tue, 02 Sep 1997 11:26:46 PDT." <199709021826.LAA13628@wile-e-coyote.cs.washington.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Sep 1997 13:13:07 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, Officially, no one has reported a successfully working i2o SMP board yet. The P6DNH is the only one I am aware of. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Tue Sep 2 14:21:09 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA28439 for smp-outgoing; Tue, 2 Sep 1997 14:21:09 -0700 (PDT) Received: from www.surfusa.com (www.surfusa.com [206.136.57.2]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA28429 for ; Tue, 2 Sep 1997 14:21:04 -0700 (PDT) Received: from ebster.surfusa.com ([206.136.57.11]) by www.surfusa.com (8.8.5/8.8.5) with SMTP id RAA01823; Tue, 2 Sep 1997 17:22:13 -0400 Message-ID: <340C832E.4CB@surfusa.com> Date: Tue, 02 Sep 1997 17:20:46 -0400 From: Eb Farris Reply-To: efarris@surfusa.com Organization: ANC - SurfUSA X-Mailer: Mozilla 3.01Gold (Win95; U) MIME-Version: 1.0 To: Steve Passe CC: mef@cs.washington.edu, smp@FreeBSD.ORG Subject: Re: smp motherboard w/ i960/i2o support References: <199709021913.NAA00552@Ilsa.StevesCafe.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Steve Passe wrote: > > Hi, > > Officially, no one has reported a successfully working i2o SMP board yet. > The P6DNH is the only one I am aware of. > > -- > Steve Passe | powered by > smp@csn.net | Symmetric MultiProcessor FreeBSD TMC has the AD6NF9 Dual PPro with I2O, Micronics has the M6DPi Dual PPro with I2O, ACER, ALR, AMI, ASUS, INTEL, Tekram are members of the I2O SIG so they may be coming out with these boards, as well. Eb Farris, ANC efarris@surfusa.com From owner-freebsd-smp Tue Sep 2 14:51:29 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA29736 for smp-outgoing; Tue, 2 Sep 1997 14:51:29 -0700 (PDT) Received: from hutcs.cs.hut.fi (hutcs.cs.hut.fi [130.233.192.7]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA29727 for ; Tue, 2 Sep 1997 14:51:25 -0700 (PDT) Received: (from apl@localhost) by hutcs.cs.hut.fi (8.8.5/8.7.3) id AAA10035; Wed, 3 Sep 1997 00:51:01 +0300 (EET DST) Message-ID: <19970903005101.00664@hutcs.cs.hut.fi> Date: Wed, 3 Sep 1997 00:51:01 +0300 From: Antti-Pekka Liedes To: freebsd-smp@freebsd.org Subject: SMP performance on P5 systems? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79 Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk The current SMP performance on my Tyan Tomcat IIID dual pentium 200MMX system is not quite what I had expected, as a simple benchmark, this is what it takes for me to compile kernel, with UP kernel: make -j3 283.19s user 32.56s system 94% cpu 5:32.83 total and with SMP: make -j3 341.23s user 134.72s system 183% cpu 4:18.80 total I was expecting the wall clock time to be less for the SMP kernel, also, there seems to be quite a lot of system overhead :-/. Of course this might be because of my relatively slow disks (I use striping on two Quantum Fireballs, both on different channel of my AHA-3940U). The benchmarks at www.freebsd.org SMP pages show a lot better performance. Is it just that ppro beats pentium so much on SMP or what? -- Antti-Pekka Liedes * apl@IRC * In two hells there's JMT 6 B 406 * apl@iki.fi * one hell too many 02150 ESPOO * apl@apocalypse.tky.hut.fi * - Lucifer +358 - 9 - 468 3121 * +358 - 40 - 5873 593 * (in God's Army) From owner-freebsd-smp Tue Sep 2 15:13:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA01219 for smp-outgoing; Tue, 2 Sep 1997 15:13:55 -0700 (PDT) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA01210 for ; Tue, 2 Sep 1997 15:13:49 -0700 (PDT) Received: (from root@localhost) by dyson.iquest.net (8.8.6/8.8.5) id RAA00487; Tue, 2 Sep 1997 17:13:39 -0500 (EST) From: "John S. Dyson" Message-Id: <199709022213.RAA00487@dyson.iquest.net> Subject: Re: SMP performance on P5 systems? In-Reply-To: <19970903005101.00664@hutcs.cs.hut.fi> from Antti-Pekka Liedes at "Sep 3, 97 00:51:01 am" To: apl@mail.cs.hut.fi (Antti-Pekka Liedes) Date: Tue, 2 Sep 1997 17:13:39 -0500 (EST) Cc: freebsd-smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Antti-Pekka Liedes said: > > The current SMP performance on my Tyan Tomcat IIID dual pentium 200MMX > system is not quite what I had expected, as a simple benchmark, this is > what it takes for me to compile kernel, with UP kernel: > make -j3 283.19s user 32.56s system 94% cpu 5:32.83 total > and with SMP: > make -j3 341.23s user 134.72s system 183% cpu 4:18.80 total > > I was expecting the wall clock time to be less for the SMP kernel, also, > there seems to be quite a lot of system overhead :-/. Of course this > might be because of my relatively slow disks (I use striping on two > Quantum Fireballs, both on different channel of my AHA-3940U). The > benchmarks at www.freebsd.org SMP pages show a lot better performance. Is > it just that ppro beats pentium so much on SMP or what? > I don't know if this is why PPros might sometimes appear to be faster in SMP than P5, but each CPU having a 2nd level cache probably has something to do with it. -- John dyson@freebsd.org jdyson@nc.com From owner-freebsd-smp Tue Sep 2 15:21:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA01572 for smp-outgoing; Tue, 2 Sep 1997 15:21:43 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA01558; Tue, 2 Sep 1997 15:21:37 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id QAA01474; Tue, 2 Sep 1997 16:21:32 -0600 (MDT) Message-Id: <199709022221.QAA01474@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Antti-Pekka Liedes cc: current@freebsd.org, smp@freebsd.org Subject: Re: serious instability with -current kernel In-reply-to: Your message of "Wed, 03 Sep 1997 01:10:54 +0300." <19970903011054.02248@hutcs.cs.hut.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Sep 1997 16:21:31 -0600 Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, A user reported: >> I'm running FreeBSD-current on my Tyan Tomcat IIID dual pentium box, and >> I've lately had serious instability problems. The latest working kernel I >> have was build on Aug 11th, all kernels after that crash randomly after >> about 2 minutes of use. There's no error message or anything, just plain [ we narrowed the problem down to screensaver lkm ] > It seems to be ok now. I made all lkms from source-tree and installed > them. Then I loaded screensaver lkm as in boot and it seems to work. I > guess my make world wasn't clean or something. A boot with it enabled will be necessary to be definative, it could be a problem with the point in time that it first comes alive. Given that the same UP kernel has no problems this is a distinct possibility. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Tue Sep 2 15:38:44 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA02433 for smp-outgoing; Tue, 2 Sep 1997 15:38:44 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA02413 for ; Tue, 2 Sep 1997 15:38:36 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id QAA01594; Tue, 2 Sep 1997 16:38:31 -0600 (MDT) Message-Id: <199709022238.QAA01594@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Antti-Pekka Liedes cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP performance on P5 systems? In-reply-to: Your message of "Wed, 03 Sep 1997 00:51:01 +0300." <19970903005101.00664@hutcs.cs.hut.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Sep 1997 16:38:31 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > > The current SMP performance on my Tyan Tomcat IIID dual pentium 200MMX > system is not quite what I had expected, as a simple benchmark, this is > what it takes for me to compile kernel, with UP kernel: > make -j3 283.19s user 32.56s system 94% cpu 5:32.83 total > and with SMP: > make -j3 341.23s user 134.72s system 183% cpu 4:18.80 total try -j16, using just 3 processes are leaving you diskbound. --- > I was expecting the wall clock time to be less for the SMP kernel, also, > there seems to be quite a lot of system overhead :-/. Of course this > might be because of my relatively slow disks (I use striping on two > Quantum Fireballs, both on different channel of my AHA-3940U). The > benchmarks at www.freebsd.org SMP pages show a lot better performance. Is > it just that ppro beats pentium so much on SMP or what? I need to see the output from "mptable -dmesg > outfile" run shortly after boot. the 3940 is a bridged PCI card and many MP BIOS's fail to probe it properly. you might be missing INTs from one channel. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Tue Sep 2 15:50:52 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA02797 for smp-outgoing; Tue, 2 Sep 1997 15:50:52 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA02792 for ; Tue, 2 Sep 1997 15:50:46 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id QAA01672; Tue, 2 Sep 1997 16:50:42 -0600 (MDT) Message-Id: <199709022250.QAA01672@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Antti-Pekka Liedes cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP performance on P5 systems? In-reply-to: Your message of "Wed, 03 Sep 1997 00:51:01 +0300." <19970903005101.00664@hutcs.cs.hut.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Sep 1997 16:50:41 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, >> The current SMP performance on my Tyan Tomcat IIID dual pentium 200MMX >> system is not quite what I had expected, as a simple benchmark, this is >> what it takes for me to compile kernel, with UP kernel: >> make -j3 283.19s user 32.56s system 94% cpu 5:32.83 total >> and with SMP: >> make -j3 341.23s user 134.72s system 183% cpu 4:18.80 total > >try -j16, using just 3 processes are leaving you diskbound. I'm sorry, I meant -j12. the point of diminishing returns is around -j12 for 2 P6s, experiment with between -j8 and -j12 to find the right number for your hardware. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Tue Sep 2 16:14:19 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id QAA03953 for smp-outgoing; Tue, 2 Sep 1997 16:14:19 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id QAA03945 for ; Tue, 2 Sep 1997 16:14:12 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id RAA01826; Tue, 2 Sep 1997 17:13:45 -0600 (MDT) Message-Id: <199709022313.RAA01826@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Hajimu UMEMOTO cc: peter@spinner.dialix.com.au, smp@FreeBSD.ORG Subject: Re: Fatal trap 12 during boot with -current In-reply-to: Your message of "Sun, 31 Aug 1997 03:43:59 +0900." <199708301844.DAA00416@peace.calm.imasy.or.jp> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Sep 1997 17:13:45 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > Yes, ipfw_mod and linux_mod were loaded. According to your > suggestion, I disabled loading ipfw_mod and reboot. Then, the kernel > was boot without any problem. :-) > > peter> Is the igmp reference an accident, or are you participating in multicast on > peter> the machine? (perhaps just routed/gated?) > > Multicast is enabled, and routed is running. Without ipfw_mod > loading, this causes no problem. as you've probably noticed we've fixed several systems suffering from lkm related problems today. have you tried rebuilding the ipfw_mod lkm and installing it manually, ie directly from its directory instead of with "make world"? one user found that that fixed another lkm for him. evidently make world didn't do it right for some reason... -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Tue Sep 2 17:31:22 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id RAA06420 for smp-outgoing; Tue, 2 Sep 1997 17:31:22 -0700 (PDT) Received: from ic.net (qmailr@srv1b.ic.net [152.160.72.20]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id RAA06415 for ; Tue, 2 Sep 1997 17:31:19 -0700 (PDT) Received: (qmail 8743 invoked from network); 3 Sep 1997 00:30:59 -0000 Received: from unknown (HELO lurch.rickl.org) (152.160.108.5) by unknown with SMTP; 3 Sep 1997 00:30:59 -0000 Received: from lurch.rickl.org (rickl@lurch.rickl.org [192.168.255.1]) by lurch.rickl.org (8.8.7/8.7.3) with ESMTP id SAA11789 for ; Tue, 2 Sep 1997 18:36:37 -0400 (EDT) Date: Tue, 2 Sep 1997 18:36:37 -0400 (EDT) From: Rick Lotoczky Reply-To: Rick Lotoczky Subject: Re: serious instability with -current kernel To: smp@freebsd.org Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I've been using the -current kernel with updates thru today (9/2) without a problem. I'm using X, and the ep network driver. The only issue I've seen is a large amount of interrupt-level overflows on the serial driver. The MB is a SuperMicro P6DOF, 64MB RAM, 2940UW scsi, and NCR '815 scsi. Rick From owner-freebsd-smp Tue Sep 2 17:56:51 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id RAA07562 for smp-outgoing; Tue, 2 Sep 1997 17:56:51 -0700 (PDT) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id RAA07547 for ; Tue, 2 Sep 1997 17:56:48 -0700 (PDT) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.6/8.8.6) with SMTP id RAA19002 for ; Tue, 2 Sep 1997 17:56:46 -0700 (PDT) Date: Tue, 2 Sep 1997 17:56:46 -0700 (PDT) From: Jaye Mathisen To: smp@freebsd.org Subject: SMP question re top: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk (different), 2 processor P6 box running SMP from 8/15/97. CPu usage on a compute bound process stays around 48-49%, la stays at 1. Is CPU usage being divided by number of CPU's? Or is it really some kind of inefficiency? From owner-freebsd-smp Tue Sep 2 20:34:19 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id UAA13526 for smp-outgoing; Tue, 2 Sep 1997 20:34:19 -0700 (PDT) Received: from pluto.plutotech.com (ken@mail.plutotech.com [206.168.67.137]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id UAA13517 for ; Tue, 2 Sep 1997 20:34:15 -0700 (PDT) Received: (from ken@localhost) by pluto.plutotech.com (8.8.5/8.8.5) id VAA05789; Tue, 2 Sep 1997 21:33:57 -0600 (MDT) From: Kenneth Merry Message-Id: <199709030333.VAA05789@pluto.plutotech.com> Subject: Re: Fatal trap 18: integer divide fault while in kernel mode In-Reply-To: <199709021822.MAA00351@Ilsa.StevesCafe.com> from Steve Passe at "Sep 2, 97 12:22:28 pm" To: smp@csn.net (Steve Passe) Date: Tue, 2 Sep 1997 21:33:56 -0600 (MDT) Cc: akira@myaw.ei.meisei-u.ac.jp, smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL28s (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Steve Passe wrote... > Hi, > > show of hands please, who is still getting "fatal trap 12" errors? I'm still getting the error, the panic is always in pmap_enter() and always occurs right at the end of the kernel boot process.. I tried rebuilding all the lkm's, no change. It even happens when I don't load any lkm's. Sorry I can't provide more details about the panic, my systems are still in a state of disarray, and I don't have remote gdb up and running... My kernel from August 13th works fine... If there's any debugging info I can provide, let me know. I would just get the system to generate a core dump, but savecore doesn't seem to like ram dumps over 64MB. (anyone know how to fix that?) Just for the record, I've got: ASUS P/I-P65UP5 w/ C-P6ND 128MB ram Adaptec 3940UW (Sony SDT-5000, Quantum XP34550W, Quantum XP39100S) 2 x SMC 10/100 cards (de driver) Hauppauge Wincast/TV, Matrox Milennium Ken -- Kenneth Merry ken@plutotech.com From owner-freebsd-smp Tue Sep 2 21:51:39 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id VAA17091 for smp-outgoing; Tue, 2 Sep 1997 21:51:39 -0700 (PDT) Received: from sendero-ppp.i-connect.net (sendero-ppp.i-Connect.Net [206.190.143.100]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id VAA17085 for ; Tue, 2 Sep 1997 21:51:33 -0700 (PDT) Received: (qmail 1076 invoked by uid 1000); 3 Sep 1997 04:51:50 -0000 Message-ID: X-Mailer: XFMail 1.2-alpha [p0] on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit MIME-Version: 1.0 In-Reply-To: <199708311909.MAA12729@tweetie-bird.cs.washington.edu> Date: Tue, 02 Sep 1997 21:51:50 -0700 (PDT) Organization: Atlas Telecom From: Simon Shapiro To: mef@cs.washington.edu Subject: Re: Open Issues on P6DNH Cc: FreeBSD-SMP@FreeBSD.ORG Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi mef@cs.washington.edu; On 31-Aug-97 you wrote: > Simon, > > What the deal with the P6DNH motherboard from Supermicro? A while > back you posted a set of problems to the smp mailing list. Are these > now resolved? Is the P6DNH MB a good one to buy if one wants to play > around with i2o? Last I left these issues, Steve said the MPTable is corrupt. I forwarded that to SuperMicro ``support'' which immediately said this is the way it should be, to stop my wild loughing, they supposedly forwarded the details to AMI. Did not hear anything since. The board is excellent, we are using many of them (we need ALL 8 PCI slots. I am working closely on this issue and hopefully in the next few weeks something will happen. Simon > > Thanks, > > Marc > > Date: Fri, 15 Aug 1997 22:55:50 -0700 (PDT) > Organization: Atlas Telecom > From: Simon Shapiro > > Hi Steve, > > Sorry for the long delay. Had to design and code a new filesystem > type, > fight a bunch of political battles, and get a misely raise after a > wonderful review (``words are cheaper than cash'', was my comment). > Hope you are doing well. > > * Although it compiles, the SoundBlaster driver is less than > functional; > cat foo.au > /dev/audio just hangs there. Works fine under UP on > the > same platform. > > * The fxp driver seems to not get interrupts at all. Everything > times out > eventually and works like a very slow polled system. This happens > IN > FRONT of the i960 bridge. The fpa (fddi) driver, the DPT driver > all work > fine. > > * If you put COM4 on irq 7 it is there, on irq 9 it is not found. A > conflict with the sb? The sb is mapped at 220. com4 is 2e8. > > * Something strange. Booted SMP, started X11, xperfmon++ insisted > that > libXaw was not there if ``shimon'' tried to tun it. Was there if > root > ran it. Permissions? Nope. I re-ompiled and re-installed it. > Now root > could see the executable (in /usr/X11R6/bin) but not ``shimon''. > Rebooted and both are fine. Never seen this one before. > > * Interrupt latency is still not so hot. The typical DPT driver in > UP > serves an interrupt in 3-55us. in SMP it takes 11-1000039us. > > * I am getting an enormous number of spurious interrupts. In UP the > typical count is 0 aborted (due to errors) and 1 for spurious (int > routine was called but hardware said ``not me''. In SMP, the > numbrs are > 0 and 62090. The 1 spurious is explainable ( a bug in > transitioning > from polled mode to interrupt mode during boot. > > If there is anything I can do to help, let me know. I have > sendero-ppp now > run smp. I have another machine I will setup tomorrow (new Brass) > which > will be the same hardware. > > Simon > --- Sincerely Yours, (Sent on 02-Sep-97, 21:17:44 by XF-Mail) Simon Shapiro Atlas Telecom Senior Architect 14355 SW Allen Blvd., Suite 130 Beaverton OR 97005 Shimon@i-Connect.Net Voice: 503.643.5559, Emergency: 503.799.2313 From owner-freebsd-smp Tue Sep 2 22:46:01 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA19840 for smp-outgoing; Tue, 2 Sep 1997 22:46:01 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA19835 for ; Tue, 2 Sep 1997 22:45:56 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id XAA03029; Tue, 2 Sep 1997 23:45:42 -0600 (MDT) Message-Id: <199709030545.XAA03029@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Simon Shapiro cc: mef@cs.washington.edu, FreeBSD-SMP@FreeBSD.ORG Subject: Re: Open Issues on P6DNH In-reply-to: Your message of "Tue, 02 Sep 1997 21:51:50 PDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Sep 1997 23:45:42 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > > What the deal with the P6DNH motherboard from Supermicro? A while > > back you posted a set of problems to the smp mailing list. Are these > > now resolved? Is the P6DNH MB a good one to buy if one wants to play > > around with i2o? > > Last I left these issues, Steve said the MPTable is corrupt. I forwarded > that to SuperMicro ``support'' which immediately said this is the way it > should be, to stop my wild loughing, they supposedly forwarded the details > to AMI. Did not hear anything since. The board is excellent, we are using > many of them (we need ALL 8 PCI slots. I am working closely on this issue > and hopefully in the next few weeks something will happen. It is my belief that the mptable is 'incorrect'. This statement could come back to embarrass me, but I would be willing to place a small wager on it. If the table is correct as is, then I have a fundimental misunderstanding of what it can look like. For the record, the part I dispute is: -- Bus: Bus ID Type 0 PCI 1 PCI 2 ISA -- I/O APICs: APIC ID Version State Address 2 0x11 usable 0xfec00000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# INT active-lo level 1 1:A 2 16 INT active-lo level 1 0:A 2 16 INT active-lo level 0 20:A 2 16 INT active-lo level 0 19:A 2 16 INT active-lo level 0 17:A 2 16 INT active-lo level 0 16:A 2 16 INT active-lo level 0 18:A 2 17 INT active-lo level 1 1:A 2 18 INT active-lo level 1 0:A 2 18 INT active-lo level 0 20:A 2 18 INT active-lo level 0 19:A 2 18 INT active-lo level 0 17:A 2 18 INT active-lo level 0 16:A 2 18 INT active-lo level 1 1:A 2 19 INT active-lo level 1 0:A 2 19 INT active-lo level 0 20:A 2 19 INT active-lo level 0 19:A 2 19 INT active-lo level 0 17:A 2 19 INT active-lo level 0 16:A 2 19 --- This mptable shows PCI INTs going to multiple APIC pins, for instance: INT active-lo level 0 20:A 2 16 INT active-lo level 0 20:A 2 18 INT active-lo level 0 20:A 2 19 PCI INT 20:A should NOT being going to 3 different APIC pins. I think what is wrong is that they mean to show 20:A, 20:B and 20:C, but they somehow mess that part up. I could have a bug in my parse code, but other similar boards parse correctly, such as this HP Netserver 5/133 (Intel XXPRESS): INT active-hi level 0 11:A 13 0 INT active-hi level 0 12:A 13 1 INT active-hi level 0 13:A 13 2 INT active-hi level 1 12:A 13 3 INT active-hi level 1 15:A 13 4 INT active-hi level 0 11:B 13 5 INT active-hi level 0 12:C 13 5 INT active-hi level 0 13:D 13 5 INT active-hi level 1 12:B 13 5 INT active-hi level 1 15:C 13 5 INT active-hi level 0 11:C 13 6 INT active-hi level 0 12:D 13 6 INT active-hi level 0 13:B 13 6 INT active-hi level 1 12:C 13 6 INT active-hi level 1 15:D 13 6 INT active-hi level 0 11:D 13 7 INT active-hi level 0 12:B 13 7 INT active-hi level 0 13:C 13 7 INT active-hi level 1 12:D 13 7 INT active-hi level 1 15:B 13 7 INT active-hi level 1 13:A 13 12 INT active-hi level 1 14:A 13 13 -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Tue Sep 2 22:57:34 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA20343 for smp-outgoing; Tue, 2 Sep 1997 22:57:34 -0700 (PDT) Received: from sendero-ppp.i-connect.net (sendero-ppp.i-Connect.Net [206.190.143.100]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id WAA20337 for ; Tue, 2 Sep 1997 22:57:30 -0700 (PDT) Received: (qmail 1952 invoked by uid 1000); 3 Sep 1997 05:57:50 -0000 Message-ID: X-Mailer: XFMail 1.2-alpha [p0] on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit MIME-Version: 1.0 In-Reply-To: <199709030545.XAA03029@Ilsa.StevesCafe.com> Date: Tue, 02 Sep 1997 22:57:50 -0700 (PDT) Organization: Atlas Telecom From: Simon Shapiro To: Steve Passe Subject: Re: Open Issues on P6DNH Cc: mef@cs.washington.edu, FreeBSD-SMP@FreeBSD.ORG Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi Steve Passe; On 03-Sep-97 you wrote: ... > --- > This mptable shows PCI INTs going to multiple APIC pins, for instance: > INT active-lo level 0 20:A 2 > 16 > INT active-lo level 0 20:A 2 > 18 > INT active-lo level 0 20:A 2 > 19 > > PCI INT 20:A should NOT being going to 3 different APIC pins. I think > what > is wrong is that they mean to show 20:A, 20:B and 20:C, but they somehow > mess that part up. I could have a bug in my parse code, but other > similar > boards parse correctly, such as this HP Netserver 5/133 (Intel XXPRESS): ... According to the expert on this matters at SuperMicro, this exactly how it should be when you have more than one PCI bus. With no ofense to anyone, in declining order of probability, here is my BELIEF: a. SuperMicro is mistaken b. SuperMicro is ``plastering our eyes'' with nonsense as they try to get out of something they have no clue about. c. Steve is wrong. Why does it work with NT? Probably, the NT driver does some sort of ``if board_id == p6dnh then reverse these pins. Should we do that too? Being true to my heritage, I will anser a question with a question: Do we want to be right about it or do we want it to work? --- Sincerely Yours, (Sent on 02-Sep-97, 22:51:34 by XF-Mail) Simon Shapiro Atlas Telecom Senior Architect 14355 SW Allen Blvd., Suite 130 Beaverton OR 97005 Shimon@i-Connect.Net Voice: 503.643.5559, Emergency: 503.799.2313 From owner-freebsd-smp Tue Sep 2 23:47:57 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA23660 for smp-outgoing; Tue, 2 Sep 1997 23:47:57 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA23652 for ; Tue, 2 Sep 1997 23:47:51 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id AAA03305; Wed, 3 Sep 1997 00:47:44 -0600 (MDT) Message-Id: <199709030647.AAA03305@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Simon Shapiro cc: mef@cs.washington.edu, FreeBSD-SMP@FreeBSD.ORG Subject: Re: Open Issues on P6DNH In-reply-to: Your message of "Tue, 02 Sep 1997 22:57:50 PDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 03 Sep 1997 00:47:44 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > According to the expert on this matters at SuperMicro, this exactly how it > should be when you have more than one PCI bus. there are 2 things wrong with that theory: 1st, with 2 busses you could have at most 2 duplicates, NOT 3. 2nd, there is a bus field to distinguish the PCI device:int pair on each bus. re-examine the HP table (an example I claim is correct), sorted by PCI device:int. Note that this shows 2 PCI busses, but NEVER more than 2 of any one device:int, and that NO pair ever claims to use the same APIC pin. NOR do both members of a pair exist on the same bus PCI bus: INT active-hi level 0 11:A 13 0 INT active-hi level 0 11:B 13 5 INT active-hi level 0 11:C 13 6 INT active-hi level 0 11:D 13 7 INT active-hi level 0 12:A 13 1 INT active-hi level 1 12:A 13 3 INT active-hi level 0 12:B 13 7 INT active-hi level 1 12:B 13 5 INT active-hi level 0 12:C 13 5 INT active-hi level 1 12:C 13 6 INT active-hi level 0 12:D 13 6 INT active-hi level 1 12:D 13 7 INT active-hi level 0 13:A 13 2 INT active-hi level 1 13:A 13 12 INT active-hi level 0 13:B 13 6 INT active-hi level 0 13:C 13 7 INT active-hi level 0 13:D 13 5 INT active-hi level 1 14:A 13 13 INT active-hi level 1 15:A 13 4 INT active-hi level 1 15:B 13 7 INT active-hi level 1 15:C 13 5 INT active-hi level 1 15:D 13 6 while the P6DNH sorted shows the same device:int connected to 3 different APIC pins in six different cases: INT active-lo level 1 0:A 2 16 INT active-lo level 1 0:A 2 18 INT active-lo level 1 0:A 2 19 INT active-lo level 1 1:A 2 16 INT active-lo level 1 1:A 2 18 INT active-lo level 1 1:A 2 19 INT active-lo level 0 16:A 2 16 INT active-lo level 0 16:A 2 18 INT active-lo level 0 16:A 2 19 INT active-lo level 0 17:A 2 16 INT active-lo level 0 17:A 2 18 INT active-lo level 0 17:A 2 19 INT active-lo level 0 18:A 2 17 INT active-lo level 0 19:A 2 16 INT active-lo level 0 19:A 2 18 INT active-lo level 0 19:A 2 19 INT active-lo level 0 20:A 2 16 INT active-lo level 0 20:A 2 18 INT active-lo level 0 20:A 2 19 In summary, they are failing to distinguish the PCI int pin, ie A,B,C,D --- > With no ofense to anyone, > in declining order of probability, here is my BELIEF: > > a. SuperMicro is mistaken > b. SuperMicro is ``plastering our eyes'' with nonsense as they try to get > out of something they have no clue about. > c. Steve is wrong. any one or more of these is possible, I would assign the same order to the possibilities. --- > Why does it work with NT? Probably, the NT driver does some sort of ``if > board_id == p6dnh then reverse these pins. MP Config Table Header: OEM ID: 'INTEL ' Product ID: '440FX ' what board id, supermicro didn't even bother to replace the generic id tags with something product specific, and I'm supposed to believe the rest of the table? Even if they did, it all falls apart when you start to plug in bridged PCI cards. Knowing the motherboard make & model is of no help then. Remember a bridged PCI card is yet another PCI bus! --- > Should we do that too? > > Being true to my heritage, I will anser a question with a question: > > Do we want to be right about it or do we want it to work? Both, in that order! The problem is that M$ doesn't even have to go out and buy one of every board made, every manufacturer sends M$ as many as M$ wants if they want the board qualified. So M$ can program around these sorts of problems as they have the hardware to test. (actually I dont think M$ even messes with NT at that level, the OEM has to do the HAL code themselves if I remember correctly.) I don't have the luxury of access to 1 of everything, my ONLY defense is to insist that the manufacturers adhere to the published standards. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Wed Sep 3 00:22:26 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA25701 for smp-outgoing; Wed, 3 Sep 1997 00:22:26 -0700 (PDT) Received: from hutcs.cs.hut.fi (hutcs.cs.hut.fi [130.233.192.7]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA25695 for ; Wed, 3 Sep 1997 00:22:21 -0700 (PDT) Received: (from apl@localhost) by hutcs.cs.hut.fi (8.8.5/8.7.3) id KAA13393; Wed, 3 Sep 1997 10:21:41 +0300 (EET DST) Message-ID: <19970903102141.63694@hutcs.cs.hut.fi> Date: Wed, 3 Sep 1997 10:21:41 +0300 From: Antti-Pekka Liedes To: Steve Passe Cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP performance on P5 systems? References: <19970903005101.00664@hutcs.cs.hut.fi> <199709022238.QAA01594@Ilsa.StevesCafe.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary=8nViIGeoXxdpn2Pv X-Mailer: Mutt 0.79 In-Reply-To: <199709022238.QAA01594@Ilsa.StevesCafe.com>; from Steve Passe on Tue, Sep 02, 1997 at 04:38:31PM -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk --8nViIGeoXxdpn2Pv Content-Type: text/plain; charset=us-ascii On Tue, Sep 02, 1997 at 04:38:31PM -0600, Steve Passe wrote: > Hi, > > > > The current SMP performance on my Tyan Tomcat IIID dual pentium 200MMX > > system is not quite what I had expected, as a simple benchmark, this is > > what it takes for me to compile kernel, with UP kernel: > > make -j3 283.19s user 32.56s system 94% cpu 5:32.83 total > > and with SMP: > > make -j3 341.23s user 134.72s system 183% cpu 4:18.80 total > > try -j16, using just 3 processes are leaving you diskbound. > -j12 or -j16 didn't give any practical differences: text data bss dec hex 1142784 98304 347076 1588164 183bc4 make -j16 349.22s user 129.98s system 189% cpu 4:12.77 total I guess it's just so heavily disk bound operation that SMP doesn't give much edge compared to UP. > --- > > I was expecting the wall clock time to be less for the SMP kernel, also, > > there seems to be quite a lot of system overhead :-/. Of course this > > might be because of my relatively slow disks (I use striping on two > > Quantum Fireballs, both on different channel of my AHA-3940U). The > > benchmarks at www.freebsd.org SMP pages show a lot better performance. Is > > it just that ppro beats pentium so much on SMP or what? > > I need to see the output from "mptable -dmesg > outfile" run shortly > after boot. the 3940 is a bridged PCI card and many MP BIOS's fail to > probe it properly. you might be missing INTs from one channel. Tomcat BIOS used to fail on this too, but the latest 4.01, which I am running, has fixed this problem. > -- > Steve Passe | powered by > smp@csn.net | Symmetric MultiProcessor FreeBSD > -- Antti-Pekka Liedes * apl@IRC * In two hells there's JMT 6 B 406 * apl@iki.fi * one hell too many 02150 ESPOO * apl@apocalypse.tky.hut.fi * - Lucifer +358 - 9 - 468 3121 * +358 - 40 - 5873 593 * (in God's Army) --8nViIGeoXxdpn2Pv Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename="mptable.out" =============================================================================== MPTable, version 2.0.14 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000f5a60 signature: '_MP_' length: 16 bytes version: 1.4 checksum: 0xc3 mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x000f5a74 signature: 'PCMP' base table length: 308 version: 1.1 checksum: 0x68 OEM ID: 'OEM00000' Product ID: 'PROD00000000' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 30 local APIC address: 0xfee00000 extended table length: 0 extended table checksum: 0 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 0 0x11 BSP, usable 5 2 1 0x07bf 1 0x11 AP, usable 5 2 1 0x07bf -- Bus: Bus ID Type 0 PCI 1 PCI 2 ISA -- I/O APICs: APIC ID Version State Address 2 0x11 usable 0xfec00000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID INT# ExtINT conforms conforms 2 0 2 0 INT conforms conforms 2 1 2 1 INT conforms conforms 2 0 2 2 INT conforms conforms 2 3 2 3 INT conforms conforms 2 4 2 4 INT conforms conforms 2 5 2 5 INT conforms conforms 2 6 2 6 INT conforms conforms 2 7 2 7 INT conforms conforms 2 8 2 8 INT conforms conforms 2 9 2 9 INT conforms conforms 2 10 2 10 INT conforms conforms 2 11 2 11 INT conforms conforms 2 12 2 12 INT conforms conforms 2 13 2 13 INT conforms conforms 2 14 2 14 INT conforms conforms 2 15 2 15 INT active-lo level 0 7:B 2 19 INT active-lo level 0 20:A 2 16 INT active-lo level 0 17:A 2 19 INT active-lo level 1 4:A 2 18 INT active-lo level 1 5:A 2 19 SMI conforms conforms 2 0 2 23 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID INT# ExtINT conforms conforms 0 0:A 255 0 NMI conforms conforms 0 0:A 255 1 ------------------------------------------------------------------------------- # SMP kernel config file options: # Required: options SMP # Symmetric MultiProcessor Kernel options APIC_IO # Symmetric (APIC) I/O # Optional (built-in defaults will work in most cases): #options NCPU=2 # number of CPUs #options NBUS=3 # number of busses #options NAPIC=1 # number of IO APICs #options NINTR=24 # number of INTs ------------------------------------------------------------------------------- dmesg output: Copyright (c) 1992-1997 FreeBSD Inc. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 3.0-CURRENT #1: Wed Sep 3 00:06:49 EEST 1997 apl@apocalypse.tky.hut.fi:/var/src/sys/compile/APOCALYPSE-SMP CPU: Pentium (586-class CPU) Origin = "GenuineIntel" Id = 0x544 Stepping=4 Features=0x8003bf real memory = 67108864 (65536K bytes) Physical memory hole(s): avail memory = 62668800 (61200K bytes) FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 0, version: 0x00030010, at 0xfee00000 cpu1 (AP): apic id: 1, version: 0x00030010, at 0xfee00000 io0 (APIC): apic id: 2, version: 0x00170011, at 0xfec00000 DEVFS: ready for devices Probing for devices on PCI bus 0: chip0: rev 0x03 on pci0.0.0 chip1: rev 0x01 on pci0.7.0 de0: rev 0x20 int a irq 19 on pci0.17.0 de0: SMC 9332BDT 21140A [10-100Mb/s] pass 2.0 de0: address 00:00:c0:30:46:f9 chip2: rev 0x02 on pci0.18.0 vga0: rev 0x01 int a irq 16 on pci0.20.0 Probing for devices on PCI bus 1: ahc0: rev 0x00 int a irq 18 on pci1.4.0 ahc0: aic7880 Channel A, SCSI Id=7, 16/255 SCBs ahc0: waiting for scsi devices to settle scbus0 at ahc0 bus 0 ahc0: target 0 Tagged Queuing Device sd0 at scbus0 target 0 lun 0 sd0: type 0 fixed SCSI 2 sd0: Direct-Access 522MB (1070496 512 byte sectors) ahc0: target 2 Tagged Queuing Device sd1 at scbus0 target 2 lun 0 sd1: type 0 fixed SCSI 2 sd1: Direct-Access 3067MB (6281856 512 byte sectors) ahc1: rev 0x00 int a irq 19 on pci1.5.0 ahc1: aic7880 Channel B, SCSI Id=7, 16/255 SCBs ahc1: waiting for scsi devices to settle scbus1 at ahc1 bus 0 ahc1: target 3 Tagged Queuing Device sd2 at scbus1 target 3 lun 0 sd2: type 0 fixed SCSI 2 sd2: Direct-Access 3067MB (6281856 512 byte sectors) ahc1: target 4 Tagged Queuing Device sd3 at scbus1 target 4 lun 0 sd3: type 0 fixed SCSI 2 sd3: Direct-Access 2006MB (4108600 512 byte sectors) Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <16 virtual consoles, flags=0x0> sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface psm0 at 0x60-0x64 irq 12 on motherboard psm0: device ID 0 fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 72065B fd0: 1.44MB 3.5in wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (atapi): , removable, iordy atapi0.0: unknown phase wdc1 not found at 0x170 sb0 at 0x220-0x22f irq 5 drq 1 on isa sb0: sbxvi0 drq 5 on isa sbxvi0: sbmidi0 at 0x330-0x331 on isa sbmidi0: npx0 on motherboard npx0: INT 16 interface apm0 on isa apm: found APM BIOS version 1.2 DEVFS: ready to run APIC_IO: routing 8254 via 8259 on pin 0 ccd0-15: Concatenated disk drivers SMP: AP CPU #1 Launched! de0: enabling Full Duplex 100baseTX port mountmsdosfs(): Warning: root directory is not a multiple of the clustersize in length =============================================================================== --8nViIGeoXxdpn2Pv-- From owner-freebsd-smp Wed Sep 3 00:40:37 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA26645 for smp-outgoing; Wed, 3 Sep 1997 00:40:37 -0700 (PDT) Received: from sendero-ppp.i-connect.net (sendero-ppp.i-Connect.Net [206.190.143.100]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id AAA26634 for ; Wed, 3 Sep 1997 00:40:33 -0700 (PDT) Received: (qmail 3270 invoked by uid 1000); 3 Sep 1997 07:40:53 -0000 Message-ID: X-Mailer: XFMail 1.2-alpha [p0] on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit MIME-Version: 1.0 In-Reply-To: <199709030647.AAA03305@Ilsa.StevesCafe.com> Date: Wed, 03 Sep 1997 00:40:53 -0700 (PDT) Organization: Atlas Telecom From: Simon Shapiro To: Steve Passe Subject: Re: Open Issues on P6DNH Cc: mef@cs.washington.edu, FreeBSD-SMP@FreeBSD.ORG Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi Steve Passe; On 03-Sep-97 you wrote: ... > > Why does it work with NT? Probably, the NT driver does some sort of > > ``if > > board_id == p6dnh then reverse these pins. > > MP Config Table Header: > OEM ID: 'INTEL ' > Product ID: '440FX ' > > what board id, supermicro didn't even bother to replace the generic id > tags with something product specific, and I'm supposed to believe the > rest of the table? >From what I understood, SuperMicro does not even have source to the BIOS. It is at AMI. Between friends, AMI slapped the BIOS together, booted DOS, ran DOOM for 10 minutes and shipped it back. I worked for a certain computer manufacturer, who shall remain nameless, but this exactly how the supercomputer division tested the boards. Oh, to test if it REALLY works, we installed Linux on the machine. We risked being fired for doing so. > Even if they did, it all falls apart when you start to plug in bridged > PCI > cards. Knowing the motherboard make & model is of no help then. > Remember > a bridged PCI card is yet another PCI bus! I put your name on the list to be ``intelectually correct''. I think you know that, Steve :-) > --- > > Should we do that too? > > > > Being true to my heritage, I will anser a question with a question: > > > > Do we want to be right about it or do we want it to work? > > Both, in that order! The problem is that M$ doesn't even have to go out > and buy one of every board made, every manufacturer sends M$ as many as > M$ wants if they want the board qualified. So M$ can program around > these sorts of problems as they have the hardware to test. (actually I > dont > think M$ even messes with NT at that level, the OEM has to do the HAL > code > themselves if I remember correctly.) If I know it right, M$ never as much as smelled this board. Someone at the nameless company hacked the HAL layer. If M$ included the hack in their CD, all others are eternally grateful. > I don't have the luxury of access to 1 of everything, my ONLY defense is > to insist that the manufacturers adhere to the published standards. The only problem we have is that if you ``insist'' on anything they will all die loughing. If I call SuperMicro and say I am returning 200 boards and ask our corporate counsel to write them a polite letter, maybe they will react. I think a posting to a good newsgroup is very useful too. As I said, I will take care of this problem in a matter satisfactory to us all. Soon. --- Sincerely Yours, (Sent on 03-Sep-97, 00:12:59 by XF-Mail) Simon Shapiro Atlas Telecom Senior Architect 14355 SW Allen Blvd., Suite 130 Beaverton OR 97005 Shimon@i-Connect.Net Voice: 503.643.5559, Emergency: 503.799.2313 From owner-freebsd-smp Wed Sep 3 00:53:09 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA27065 for smp-outgoing; Wed, 3 Sep 1997 00:53:09 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA27057 for ; Wed, 3 Sep 1997 00:53:07 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id BAA03651; Wed, 3 Sep 1997 01:52:59 -0600 (MDT) Message-Id: <199709030752.BAA03651@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Simon Shapiro cc: mef@cs.washington.edu, FreeBSD-SMP@FreeBSD.ORG Subject: Re: Open Issues on P6DNH In-reply-to: Your message of "Wed, 03 Sep 1997 00:40:53 PDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 03 Sep 1997 01:52:59 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, --- First, I'm not getting "defensive" or "pissed" about this discussion, I'm just trying to strongly support my point. (perhaps I am a bit pissed at supermicro...) --- > The only problem we have is that if you ``insist'' on anything they will > all die loughing. What ever works, as long as they die! -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Wed Sep 3 04:49:58 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id EAA07381 for smp-outgoing; Wed, 3 Sep 1997 04:49:58 -0700 (PDT) Received: from fang.cs.sunyit.edu (perlsta@fang.cs.sunyit.edu [192.52.220.66]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id EAA07375 for ; Wed, 3 Sep 1997 04:49:53 -0700 (PDT) Received: from localhost (perlsta@localhost) by fang.cs.sunyit.edu (8.8.5/8.7.3) with SMTP id LAA17715; Wed, 3 Sep 1997 11:49:52 GMT Date: Wed, 3 Sep 1997 11:49:52 +0000 (GMT) From: Alfred Perlstein To: Antti-Pekka Liedes cc: Steve Passe , freebsd-smp@FreeBSD.ORG Subject: Re: SMP performance on P5 systems? In-Reply-To: <19970903102141.63694@hutcs.cs.hut.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk What kind of drive technology are you using? non-busmastering IDE will kill your perforance won't it? > > > The current SMP performance on my Tyan Tomcat IIID dual pentium 200MMX > > > system is not quite what I had expected, as a simple benchmark, this is > > > what it takes for me to compile kernel, with UP kernel: > > > make -j3 283.19s user 32.56s system 94% cpu 5:32.83 total > > > and with SMP: > > > make -j3 341.23s user 134.72s system 183% cpu 4:18.80 total > > > > try -j16, using just 3 processes are leaving you diskbound. > > > > -j12 or -j16 didn't give any practical differences: > text data bss dec hex > 1142784 98304 347076 1588164 183bc4 > make -j16 349.22s user 129.98s system 189% cpu 4:12.77 total > > I guess it's just so heavily disk bound operation that SMP doesn't give > much edge compared to UP. From owner-freebsd-smp Wed Sep 3 10:10:05 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id KAA23862 for smp-outgoing; Wed, 3 Sep 1997 10:10:05 -0700 (PDT) Received: from bagpuss.visint.co.uk (bagpuss.visint.co.uk [194.207.134.1]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id KAA23825 for ; Wed, 3 Sep 1997 10:09:59 -0700 (PDT) Received: from dylan.visint.co.uk (dylan.visint.co.uk [194.207.134.180]) by bagpuss.visint.co.uk (8.7.5/8.7.3) with SMTP id SAA17266 for ; Wed, 3 Sep 1997 18:09:54 +0100 (BST) Date: Wed, 3 Sep 1997 18:09:58 +0100 (BST) From: Stephen Roome To: smp@freebsd.org Subject: rndcontrol Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Okay, this is a really minor thing, but all these irq's have been assigned to numbers which don't really work in the case of UP. This means I can't use rndcontrol -s 18 -s 19, to use the SCSI or video card as random sources. My point.. well, is the limitation rndcontrol or ioctl? and does it matter much ?? (I'm -still- running 3.0-970618-SNAP) -- Steve Roome - Vision Interactive Ltd. Tel:+44(0)117 9730597 Home:+44(0)976 241342 WWW: http://dylan.visint.co.uk/ From owner-freebsd-smp Wed Sep 3 11:04:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA27246 for smp-outgoing; Wed, 3 Sep 1997 11:04:06 -0700 (PDT) Received: from tasogare.imasy.or.jp (root@tasogare.imasy.or.jp [202.227.24.5]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA27209 for ; Wed, 3 Sep 1997 11:03:58 -0700 (PDT) Received: (from ume@localhost) by tasogare.imasy.or.jp (8.8.7+2.7Wbeta7/3.4W4-96030215) with UUCP id CAA00535; Thu, 4 Sep 1997 02:38:19 +0900 (JST) Received: from peace.calm.imasy.or.jp (root@peace.calm.imasy.or.jp [158.214.107.233]) by chaos.calm.imasy.or.jp (8.8.7/3.6Wbeta6-CHAOS1.5) with ESMTP id CAA07225; Thu, 4 Sep 1997 02:37:13 +0900 (JST) Received: from localhost (ume@localhost [127.0.0.1]) by peace.calm.imasy.or.jp (8.8.7/3.6Wbeta6-CALM1.0) with ESMTP id CAA00277; Thu, 4 Sep 1997 02:37:06 +0900 (JST) Message-Id: <199709031737.CAA00277@peace.calm.imasy.or.jp> To: smp@csn.net Cc: peter@spinner.dialix.com.au, smp@FreeBSD.ORG Subject: Re: Fatal trap 12 during boot with -current In-Reply-To: Your message of "Tue, 02 Sep 1997 17:13:45 -0600" <199709022313.RAA01826@Ilsa.StevesCafe.com> References: <199709022313.RAA01826@Ilsa.StevesCafe.com> X-Mailer: Mew version 1.90 on XEmacs 20.3 (Bratislava) X-PGP-Fingerprint: 6B 0C 53 FC 5D D0 37 91 05 D0 B3 EF 36 9B 6A BC X-URL: http://www.imasy.or.jp/~ume/ Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Date: Thu, 04 Sep 1997 02:37:06 +0900 From: Hajimu UMEMOTO X-Dispatcher: imput version 970830 Lines: 21 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, >>>>> On Tue, 02 Sep 1997 17:13:45 -0600, Steve Passe said: smp> as you've probably noticed we've fixed several systems suffering from smp> lkm related problems today. have you tried rebuilding the ipfw_mod lkm smp> and installing it manually, ie directly from its directory instead smp> of with "make world"? one user found that that fixed another lkm for him. smp> evidently make world didn't do it right for some reason... I built lkms during `make world'. I wish to try that method, but... I've tried with the kernel cvsuped at Sep 3 and Sep 4. Although no lkm module is loaded, when accessing network, the kernel causes panic frequently. The UP kernel seems to have no problem. I'm using vx driver for 3C905. -- Hajimu UMEMOTO @ Internet Mutual Aid Society Yokohama, Japan ume@imasy.or.jp ume@iabs.hitachi.co.jp http://www.imasy.or.jp/~ume/ From owner-freebsd-smp Wed Sep 3 15:22:52 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA12063 for smp-outgoing; Wed, 3 Sep 1997 15:22:52 -0700 (PDT) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA12058 for ; Wed, 3 Sep 1997 15:22:47 -0700 (PDT) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.6/8.8.6) with SMTP id PAA18608 for ; Wed, 3 Sep 1997 15:22:45 -0700 (PDT) Date: Wed, 3 Sep 1997 15:22:45 -0700 (PDT) From: Jaye Mathisen To: smp@freebsd.org Subject: Update on panic. Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Doesn't happen with a uniprocessor kernel on the same hardware, so seems definitely tied to SMP. From owner-freebsd-smp Thu Sep 4 10:21:29 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id KAA12326 for smp-outgoing; Thu, 4 Sep 1997 10:21:29 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id KAA12315; Thu, 4 Sep 1997 10:21:13 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id LAA09408; Thu, 4 Sep 1997 11:20:37 -0600 (MDT) Message-Id: <199709041720.LAA09408@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: dyson@FreeBSD.ORG cc: current@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: 3.0/SMP panic In-reply-to: Your message of "Thu, 04 Sep 1997 00:31:20 CDT." <199709040531.AAA03246@dyson.iquest.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 04 Sep 1997 11:20:37 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk John, > I have been having the same panic at pmap_enter (note that it ISN'T a > pmap_enter bug.) At least I can reproduce it, and it appears to be excited > by simple (kernel) code mods or movement. There is a problem lurking, and I > have spent the last 4-5 days trying to track it down. I am now more > methodically trying to ferret the problem out, and am trying to fix it ASAP. > It has held up my deliverables and it NEEDS to be fixed!!! :-(. one user reports this happens to him with SMP but not UP. In your setup where you can reproduce it, is it SMP only or UP also? -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Thu Sep 4 10:32:11 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id KAA13016 for smp-outgoing; Thu, 4 Sep 1997 10:32:11 -0700 (PDT) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id KAA13009; Thu, 4 Sep 1997 10:32:05 -0700 (PDT) Received: (from root@localhost) by dyson.iquest.net (8.8.6/8.8.5) id MAA01880; Thu, 4 Sep 1997 12:31:53 -0500 (EST) From: "John S. Dyson" Message-Id: <199709041731.MAA01880@dyson.iquest.net> Subject: Re: 3.0/SMP panic In-Reply-To: <199709041720.LAA09408@Ilsa.StevesCafe.com> from Steve Passe at "Sep 4, 97 11:20:37 am" To: smp@csn.net (Steve Passe) Date: Thu, 4 Sep 1997 12:31:52 -0500 (EST) Cc: dyson@FreeBSD.ORG, current@FreeBSD.ORG, smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Steve Passe said: > John, > > > I have been having the same panic at pmap_enter (note that it ISN'T a > > pmap_enter bug.) At least I can reproduce it, and it appears to be excited > > by simple (kernel) code mods or movement. There is a problem lurking, and I > > have spent the last 4-5 days trying to track it down. I am now more > > methodically trying to ferret the problem out, and am trying to fix it ASAP. > > It has held up my deliverables and it NEEDS to be fixed!!! :-(. > > one user reports this happens to him with SMP but not UP. In your setup > where you can reproduce it, is it SMP only or UP also? > I haven't been able to reproduce it in UP, but I really don't think that it is an SMP problem. However, that is only my "feeling." -- John dyson@freebsd.org jdyson@nc.com From owner-freebsd-smp Thu Sep 4 11:09:15 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA14410 for smp-outgoing; Thu, 4 Sep 1997 11:09:15 -0700 (PDT) Received: from apolo.lsi.usp.br (mario@apolo.lsi.usp.br [143.107.161.199]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA14404 for ; Thu, 4 Sep 1997 11:09:09 -0700 (PDT) Received: from localhost (mario@localhost) by apolo.lsi.usp.br (8.8.5/8.8.5) with SMTP id CAA02715 for ; Fri, 5 Sep 1997 02:07:51 -0300 Date: Fri, 5 Sep 1997 02:07:51 -0300 (EST) From: Mario Donato Marino To: smp@freebsd.org Subject: freebsd Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk unsubscribe smp mario@lsi.usp.br From owner-freebsd-smp Thu Sep 4 12:11:34 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id MAA17438 for smp-outgoing; Thu, 4 Sep 1997 12:11:34 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id MAA17423 for ; Thu, 4 Sep 1997 12:11:28 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id NAA09822; Thu, 4 Sep 1997 13:11:24 -0600 (MDT) Message-Id: <199709041911.NAA09822@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: smp@freebsd.org cc: Fred Clark Subject: the majordomo reply for SMP Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 04 Sep 1997 13:11:24 -0600 Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, Fred Clark pointed out that the reply sent by majordomo when subscribing to SMP is sorely out of date, among other things describing the old pre-merge method of getting the source. Please ignore this info. The new message will look something like: FREEBSD-SMP Technical discussions This is a forum for technical discussions related to the FreeBSD Symmetric MultiProcessor kernel. This is the primary technical mailing list. It is for individuals actively working on FreeBSD SMP, to bring up problems or discuss alternative solutions. Individuals interested in following the technical discussion are also welcome. Be aware that FreeBSD SMP is currently BETA level software, and should NOT be used for mission critical applications. In addition it is subject to all the caveats of FreeBSD current, anyone using the SMP kernel should also be subscribed to current@FreeBSD.org. Further information on installation and maintaince can be found at: http://www.freebsd.org/~fsmp/SMP/SMP.html -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Thu Sep 4 13:09:52 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id NAA20518 for smp-outgoing; Thu, 4 Sep 1997 13:09:52 -0700 (PDT) Received: from onyx.atipa.com (user1807@ns.atipa.com [208.128.22.10]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id NAA20511 for ; Thu, 4 Sep 1997 13:09:47 -0700 (PDT) Received: (qmail-queue invoked by uid 1018); 4 Sep 1997 20:13:12 -0000 Date: Thu, 4 Sep 1997 14:13:10 -0600 (MDT) From: Atipa X-Sender: freebsd@dot.ishiboo.com To: Steve Passe cc: smp@FreeBSD.ORG Subject: Warning: DO NOT reflash GA-686DX boards In-Reply-To: <199709041911.NAA09822@Ilsa.StevesCafe.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Anybody using GA-686DX motherboards should be aware of a problem with the most recent BIOS that kills the performance of the 2nd processor. Well, it doesn't kill it, but it wounds it to the point you would be better off with a 386sx-12 for your 2nd processor... I believe the most recent GOOD BIOS is 1.23. "If it ain't broke, don't fix it." Cordially, Kevin Below is a forward from the Linux world relating to the problem: >There definitely is a problem with the Gigabyte GA-686DX motherboard >in the SMP mode under flavors of Unix. We were alerted to the same >problem under Unixware. > >As Linus has alerted - we have been told by Gigabyte that it can be >solved through a BIOS update. Unfortunately, it hasn't come - we have >been waiting at least a month for it. > >BTW, the design group of the board in Taiwan was notified of your >problems From owner-freebsd-smp Thu Sep 4 13:21:26 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id NAA21040 for smp-outgoing; Thu, 4 Sep 1997 13:21:26 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id NAA21028 for ; Thu, 4 Sep 1997 13:21:19 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id OAA10238; Thu, 4 Sep 1997 14:21:05 -0600 (MDT) Message-Id: <199709042021.OAA10238@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: "John S. Dyson" cc: smp@FreeBSD.ORG Subject: Re: 3.0/SMP panic In-reply-to: Your message of "Thu, 04 Sep 1997 12:31:52 CDT." <199709041731.MAA01880@dyson.iquest.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 04 Sep 1997 14:21:05 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, We have gotten most SMP systems running now, one recent hurdle was lkms that got out of sync with the kernel proper. The symptom was panic during boot, or possibly when a screensaver lkm activated. ipfw_mod was also shown to be a problem. The solution is to sup current source for the lkms, rebuild & install them. We still have at least one fundimental bug affecting a small number of systems: Fatal trap 12 during boot with -current. This bug has so far only been seen under SMP (is this true?). It appears to be very dependant on the specific system configuration. The following is a roundup of reports from various users. Unless your working on this problem you probably don't want to read further. --- Kenneth Merry : > By any chance do you have more than 64MB in your machine and >options MAXMEM=... in your kernel config file? > > I did, and I had panics very much like that (in pmap_enter) >immediately on boot. When I took the MAXMEM line out (I've got 128MB), >things worked just fine... I'm still not sure why, though. > ... > I found the problem. At first I suspected the sound driver, but > the problem really turned out to be: > > options "MAXMEM=(128*1024)" --- Jaye Mathisen : I was using M$ Inetload 2.0 to simulate a bunch of mail users. IT was running fine for a few minutes, then died horribly with: Fatal trap 12: page fault while in kernel mode cpuid = 1 lapic.id = 33554432 current process = Idle mp_lock = 01000003 interrupt mask = net tty bio <- SMP: XXX Stopped at _pmap_enter+0xa7: and some other stuff. The traceback is not too long, but I don't have any good way to type it all in. It goes like: _pmap_enter _vm_fault Trap_pfault _trap _zalloc _pmap_insert_entry _pmap_enter _kmem_alloc _in_pcballoc _Tcp_attach _tcp_usr_attack _sonewconn _tcp_inut _ip_input _ipintr swi_net_next It is trivially reproducible at least on my hardware. --- Akira Watanabe : The kernel (suped yesterday) causes a panic. Fatal trap 18: integer divide fault while in kernel mode cpuid = 0 lapic.id = 16777216 instruction pointer = 0x8:0xf01bc794 stack pointer = 0x10:0xf4cabc84 frame pointer = 0x10:0xf4cabcd0 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 236 (ftpd) mp_lock = 00000003 interrupt mask = <- SMP: XXX trap number = 18 panic: integer divide fault cpuid 0 boot() called on cpu#0 syncing disks... 11 11 8 2 done Here is a stack trace. # gdb -k kernel /var/crash/vmcore.0 GDB is free software and you are welcome to distribute copies of it under certain conditions; type "show copying" to see the conditions. There is absolutely no warranty for GDB; type "show warranty" for details. GDB 4.16 (i386-unknown-freebsd), Copyright 1996 Free Software Foundation, Inc... IdlePTD 24d000 current pcb at 1f9608 panic: integer divide fault #0 boot (howto=256) at ../../kern/kern_shutdown.c:289 289 dumppcb.pcb_cr3 = rcr3(); (kgdb) where #0 boot (howto=256) at ../../kern/kern_shutdown.c:289 #1 0xf0118e36 in panic (fmt=0xf01ccbea "integer divide fault") at ../../kern/kern_shutdown.c:416 #2 0xf01cd86f in trap_fatal (frame=0xf4cabc48) at ../../i386/i386/trap.c:806 #3 0xf01cd072 in trap (frame={tf_es = -256049136, tf_ds = 131088, tf_edi = -256677376, tf_esi = 0, tf_ebp = -188039984, tf_isp = -188040080, tf_ebx = 0, tf_edx = 0, tf_ecx = 4096, tf_eax = 4096, tf_trapno = 18, tf_err = 0, tf_eip = -266614892, tf_cs = 8, tf_eflags = 66118, tf_esp = 0, tf_ss = 3}) at ../../i386/i386/trap.c:487 #4 0xf01bc794 in vnode_pager_haspage (object=0xf0bdb800, pindex=0, before=0xf4cabd34, after=0xf4cabd30) at ../../vm/vnode_pager.c:231 #5 0xf01bbcff in vm_pager_has_page (object=0xf0bdb800, offset=0, before=0xf4cabd34, after=0xf4cabd30) at ../../vm/vm_pager.c:205 #6 0xf01b2e05 in vm_fault_additional_pages (m=0xf04ef7c4, rbehind=3, rahead=4, marray=0xf4cabdd0, reqpage=0xf4cabda4) at ../../vm/vm_fault.c:1100 #7 0xf01b21c0 in vm_fault (map=0xf0bd9300, vaddr=134385664, fault_type=1 '\001', fault_flags=0) at ../../vm/vm_fault.c:414 #8 0xf01cd23a in trap_pfault (frame=0xf4cabe50, usermode=0) at ../../i386/i386/trap.c:681 #9 0xf01ccf47 in trap (frame={tf_es = 134348816, tf_ds = 134348816, tf_edi = -259133440, tf_esi = 134385664, tf_ebp = -188039496, tf_isp = -188039560, tf_ebx = 2048, tf_edx = 134387712, tf_ecx = 512, tf_eax = -188047360, tf_trapno = 12, tf_err = 0, tf_eip = -266551371, tf_cs = 8, tf_eflags = 66054, tf_esp = -188039368, tf_ss = -188039376}) at ../../i386/i386/trap.c:339 #10 0xf01cbfb5 in generic_copyin () #11 0xf012cf6f in sosend (so=0xf0bddc00, addr=0x0, uio=0xf4cabf38, top=0x0, control=0x0, flags=0, p=0xf0bb9600) at ../../kern/uipc_socket.c:449 #12 0xf0122ec8 in soo_write (fp=0xf0bdcd40, uio=0xf4cabf38, cred=0xf0bd8b00) at ../../kern/sys_socket.c:78 #13 0xf0120884 in write (p=0xf0bb9600, uap=0xf4cabf94, retval=0xf4cabf84) at ../../kern/sys_generic.c:268 #14 0xf01cdacb in syscall (frame={tf_es = 39, tf_ds = 39, tf_edi = 134385664, tf_esi = 2256, tf_ebp = -272642012, tf_isp = -188039196, tf_ebx = 6, tf_edx = 5, tf_ecx = 1, tf_eax = 4, tf_trapno = 22, tf_err = 7, tf_eip = 135028641, tf_cs = 31, tf_eflags = 531, tf_esp = -272642064, tf_ss = 39}) at ../../i386/i386/trap.c:953 #15 0x80c5fa1 in ?? () #16 0x3658 in ?? () #17 0x86fb in ?? () #18 0x2045 in ?? () #19 0x1096 in ?? () (kgdb) --- Hajimu UMEMOTO : Sept 1: > Yes, ipfw_mod and linux_mod were loaded. According to your > suggestion, I disabled loading ipfw_mod and reboot. Then, the kernel > was boot without any problem. :-) Sept 4: > I built lkms during `make world'. I wish to try that method, but... > I've tried with the kernel cvsuped at Sep 3 and Sep 4. Although no > lkm module is loaded, when accessing network, the kernel causes panic > frequently. The UP kernel seems to have no problem. I'm using vx > driver for 3C905. --- Tom Bartol : Over the last several days starting from world/kernel of 8/28 and even on world/kernel as of last night (9/2) I get crashes that haven't left me with any useful info. All the crashes have occured while composing e-mail from within pine. My /var/mail is an NFSv3 mounted fs served by an Auspex NS-7000 over 100/BT (nice!). The system in question is a Dell XPS-P133c (i.e. P5/133) with 128MB, Adaptec 2940U, and 3Com 3C595 100/BT. I've been running the same world/kernel on my home system with no trouble (but no NFS or network card either). Curiously, I composed this e-mail on the unstable system with no trouble. All the crashes consistently occured while composing mail within a few minutes after logging in. --- From: randyd To: smp@csn.net Subject: SMP / LKM update Greetings, Just a quick update... I cvsupped new code at about 7:30 CST yesterday and did "make cleandepend && make world". This AM I built a fresh SMP kernel and rebooted the machine. I didn't start an X session though, I waited for the 'daemon' screen saver to "kick in". When it did, I got a screen full of... Oops I'm on cpu#1, I need to be on Cpu#0 -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Thu Sep 4 14:46:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA25268 for smp-outgoing; Thu, 4 Sep 1997 14:46:13 -0700 (PDT) Received: from gauss.ee.calpoly.edu (root@gauss.ee.calpoly.edu [129.65.26.111]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA25263 for ; Thu, 4 Sep 1997 14:46:10 -0700 (PDT) Received: (from kwayman@localhost) by gauss.ee.calpoly.edu (8.8.5/ELEE) id OAA02544 for smp@freebsd.org; Thu, 4 Sep 1997 14:40:51 -0700 (PDT) From: Kyle Wayman Message-Id: <199709042140.OAA02544@gauss.ee.calpoly.edu> Subject: A tough choice... To: smp@freebsd.org Date: Thu, 4 Sep 1997 14:40:51 -0700 (PDT) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello. We are receiving a new server at the University that I work for. It is supposedly a Dual Pentium II with 256MB of RAM and various other sundries. This is to be a file/web/other server for about 900 students. Being a System Administrator of Unix systems I would like to keep it as a Unix server. Is FreeBSD SMP ready for this sort of task? If no, how long should I wait? I thank anyone for any input and/or refferences to other solutions. -- Kyle Wayman System Administrator Cal Poly - EE Department -- P.S. Thanks for the advice Steve :) From owner-freebsd-smp Thu Sep 4 16:18:23 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id QAA29102 for smp-outgoing; Thu, 4 Sep 1997 16:18:23 -0700 (PDT) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id QAA29097 for ; Thu, 4 Sep 1997 16:18:13 -0700 (PDT) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.6/8.8.6) with SMTP id QAA09584; Thu, 4 Sep 1997 16:17:54 -0700 (PDT) Date: Thu, 4 Sep 1997 16:17:53 -0700 (PDT) From: Jaye Mathisen To: Steve Passe cc: "John S. Dyson" , smp@FreeBSD.ORG Subject: Re: 3.0/SMP panic In-Reply-To: <199709042021.OAA10238@Ilsa.StevesCafe.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Recent kernel changes (maybe the tcp stuff garrett did relating to software interrupts? (I have a DPT controller), seems to have fixed it, The same load that was crashing it yesterday after supping and hbuilding and booting new kernel is running fine. On Thu, 4 Sep 1997, Steve Passe wrote: > Hi, > > We have gotten most SMP systems running now, one recent hurdle was lkms > that got out of sync with the kernel proper. The symptom was panic during > boot, or possibly when a screensaver lkm activated. ipfw_mod was also shown > to be a problem. > The solution is to sup current source for the lkms, rebuild & install them. > > We still have at least one fundimental bug affecting a small number of > systems: Fatal trap 12 during boot with -current. > > This bug has so far only been seen under SMP (is this true?). It appears to > be very dependant on the specific system configuration. The following is > a roundup of reports from various users. Unless your working on this > problem you probably don't want to read further. > > --- > Kenneth Merry : > > > By any chance do you have more than 64MB in your machine and > >options MAXMEM=... in your kernel config file? > > > > I did, and I had panics very much like that (in pmap_enter) > >immediately on boot. When I took the MAXMEM line out (I've got 128MB), > >things worked just fine... I'm still not sure why, though. > > ... > > I found the problem. At first I suspected the sound driver, but > > the problem really turned out to be: > > > > options "MAXMEM=(128*1024)" > > > --- > Jaye Mathisen : > > I was using M$ Inetload 2.0 to simulate a bunch of mail users. IT was > running fine for a few minutes, then died horribly with: > > Fatal trap 12: page fault while in kernel mode > cpuid = 1 > lapic.id = 33554432 > > current process = Idle > mp_lock = 01000003 > > interrupt mask = net tty bio <- SMP: XXX > > Stopped at _pmap_enter+0xa7: > > > and some other stuff. > > The traceback is not too long, but I don't have any good way to type it > all in. > > It goes like: > > _pmap_enter > _vm_fault > Trap_pfault > _trap > _zalloc > _pmap_insert_entry > _pmap_enter > _kmem_alloc > _in_pcballoc > _Tcp_attach > _tcp_usr_attack > _sonewconn > _tcp_inut > _ip_input > _ipintr > swi_net_next > > > It is trivially reproducible at least on my hardware. > > > --- > Akira Watanabe : > > The kernel (suped yesterday) causes a panic. > > Fatal trap 18: integer divide fault while in kernel mode > cpuid = 0 > lapic.id = 16777216 > instruction pointer = 0x8:0xf01bc794 > stack pointer = 0x10:0xf4cabc84 > frame pointer = 0x10:0xf4cabcd0 > code segment = base 0x0, limit 0xfffff, type 0x1b > = DPL 0, pres 1, def32 1, gran 1 > processor eflags = interrupt enabled, resume, IOPL = 0 > current process = 236 (ftpd) > mp_lock = 00000003 > interrupt mask = <- SMP: XXX > trap number = 18 > panic: integer divide fault > cpuid 0 > boot() called on cpu#0 > > syncing disks... 11 11 8 2 done > > Here is a stack trace. > > # gdb -k kernel /var/crash/vmcore.0 > GDB is free software and you are welcome to distribute copies of it > under certain conditions; type "show copying" to see the conditions. > There is absolutely no warranty for GDB; type "show warranty" for details. > GDB 4.16 (i386-unknown-freebsd), > Copyright 1996 Free Software Foundation, Inc... > IdlePTD 24d000 > current pcb at 1f9608 > panic: integer divide fault > #0 boot (howto=256) at ../../kern/kern_shutdown.c:289 > 289 dumppcb.pcb_cr3 = rcr3(); > (kgdb) where > #0 boot (howto=256) at ../../kern/kern_shutdown.c:289 > #1 0xf0118e36 in panic (fmt=0xf01ccbea "integer divide fault") > at ../../kern/kern_shutdown.c:416 > #2 0xf01cd86f in trap_fatal (frame=0xf4cabc48) at ../../i386/i386/trap.c:806 > #3 0xf01cd072 in trap (frame={tf_es = -256049136, tf_ds = 131088, > tf_edi = -256677376, tf_esi = 0, tf_ebp = -188039984, > tf_isp = -188040080, tf_ebx = 0, tf_edx = 0, tf_ecx = 4096, > tf_eax = 4096, tf_trapno = 18, tf_err = 0, tf_eip = -266614892, > tf_cs = 8, tf_eflags = 66118, tf_esp = 0, tf_ss = 3}) > at ../../i386/i386/trap.c:487 > #4 0xf01bc794 in vnode_pager_haspage (object=0xf0bdb800, pindex=0, > before=0xf4cabd34, after=0xf4cabd30) at ../../vm/vnode_pager.c:231 > #5 0xf01bbcff in vm_pager_has_page (object=0xf0bdb800, offset=0, > before=0xf4cabd34, after=0xf4cabd30) at ../../vm/vm_pager.c:205 > #6 0xf01b2e05 in vm_fault_additional_pages (m=0xf04ef7c4, rbehind=3, > rahead=4, marray=0xf4cabdd0, reqpage=0xf4cabda4) > at ../../vm/vm_fault.c:1100 > #7 0xf01b21c0 in vm_fault (map=0xf0bd9300, vaddr=134385664, > fault_type=1 '\001', fault_flags=0) at ../../vm/vm_fault.c:414 > #8 0xf01cd23a in trap_pfault (frame=0xf4cabe50, usermode=0) > at ../../i386/i386/trap.c:681 > #9 0xf01ccf47 in trap (frame={tf_es = 134348816, tf_ds = 134348816, > tf_edi = -259133440, tf_esi = 134385664, tf_ebp = -188039496, > tf_isp = -188039560, tf_ebx = 2048, tf_edx = 134387712, tf_ecx = 512, > tf_eax = -188047360, tf_trapno = 12, tf_err = 0, tf_eip = -266551371, > tf_cs = 8, tf_eflags = 66054, tf_esp = -188039368, tf_ss = -188039376}) > at ../../i386/i386/trap.c:339 > #10 0xf01cbfb5 in generic_copyin () > #11 0xf012cf6f in sosend (so=0xf0bddc00, addr=0x0, uio=0xf4cabf38, top=0x0, > control=0x0, flags=0, p=0xf0bb9600) at ../../kern/uipc_socket.c:449 > #12 0xf0122ec8 in soo_write (fp=0xf0bdcd40, uio=0xf4cabf38, cred=0xf0bd8b00) > at ../../kern/sys_socket.c:78 > #13 0xf0120884 in write (p=0xf0bb9600, uap=0xf4cabf94, retval=0xf4cabf84) > at ../../kern/sys_generic.c:268 > #14 0xf01cdacb in syscall (frame={tf_es = 39, tf_ds = 39, tf_edi = 134385664, > tf_esi = 2256, tf_ebp = -272642012, tf_isp = -188039196, tf_ebx = 6, > tf_edx = 5, tf_ecx = 1, tf_eax = 4, tf_trapno = 22, tf_err = 7, > tf_eip = 135028641, tf_cs = 31, tf_eflags = 531, tf_esp = -272642064, > tf_ss = 39}) at ../../i386/i386/trap.c:953 > #15 0x80c5fa1 in ?? () > #16 0x3658 in ?? () > #17 0x86fb in ?? () > #18 0x2045 in ?? () > #19 0x1096 in ?? () > (kgdb) > > > --- > Hajimu UMEMOTO : > > Sept 1: > > Yes, ipfw_mod and linux_mod were loaded. According to your > > suggestion, I disabled loading ipfw_mod and reboot. Then, the kernel > > was boot without any problem. :-) > > Sept 4: > > I built lkms during `make world'. I wish to try that method, but... > > I've tried with the kernel cvsuped at Sep 3 and Sep 4. Although no > > lkm module is loaded, when accessing network, the kernel causes panic > > frequently. The UP kernel seems to have no problem. I'm using vx > > driver for 3C905. > > > --- > Tom Bartol : > > Over the last several days starting from world/kernel of 8/28 and even on > world/kernel as of last night (9/2) I get crashes that haven't left me > with any useful info. All the crashes have occured while composing e-mail > from within pine. My /var/mail is an NFSv3 mounted fs served by an Auspex > NS-7000 over 100/BT (nice!). The system in question is a Dell XPS-P133c > (i.e. P5/133) with 128MB, Adaptec 2940U, and 3Com 3C595 100/BT. I've > been running the same world/kernel on my home system with no trouble (but > no NFS or network card either). Curiously, I composed this e-mail on the > unstable system with no trouble. All the crashes consistently occured > while composing mail within a few minutes after logging in. > > > --- > From: randyd > To: smp@csn.net > Subject: SMP / LKM update > > Greetings, > > Just a quick update... > > I cvsupped new code at about 7:30 CST yesterday and did "make > cleandepend && make > world". This AM I built a fresh SMP kernel and rebooted the machine. I > didn't > start an X session though, I waited for the 'daemon' screen saver to > "kick in". > When it did, I got a screen full of... > Oops I'm on cpu#1, I need to be on Cpu#0 > > > -- > Steve Passe | powered by > smp@csn.net | Symmetric MultiProcessor FreeBSD > > From owner-freebsd-smp Thu Sep 4 16:23:00 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id QAA29347 for smp-outgoing; Thu, 4 Sep 1997 16:23:00 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id QAA29336 for ; Thu, 4 Sep 1997 16:22:56 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id RAA11106; Thu, 4 Sep 1997 17:22:43 -0600 (MDT) Message-Id: <199709042322.RAA11106@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Jaye Mathisen cc: "John S. Dyson" , smp@FreeBSD.ORG Subject: Re: 3.0/SMP panic In-reply-to: Your message of "Thu, 04 Sep 1997 16:17:53 PDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 04 Sep 1997 17:22:42 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > Recent kernel changes (maybe the tcp stuff garrett did relating to > software interrupts? (I have a DPT controller), seems to have fixed it, > The same load that was crashing it yesterday after supping and hbuilding > and booting new kernel is running fine. unfortunately not, I have a current sys tree from about 2 hours ago and I can get the panic by loading the daemon screen saver. it probably just moved things around in a way that hides the problem for you... -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Thu Sep 4 17:03:51 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id RAA02660 for smp-outgoing; Thu, 4 Sep 1997 17:03:51 -0700 (PDT) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id RAA02654 for ; Thu, 4 Sep 1997 17:03:47 -0700 (PDT) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.6/8.8.6) with SMTP id RAA19547; Thu, 4 Sep 1997 17:03:21 -0700 (PDT) Date: Thu, 4 Sep 1997 17:03:20 -0700 (PDT) From: Jaye Mathisen To: Steve Passe cc: "John S. Dyson" , smp@FreeBSD.ORG Subject: Re: 3.0/SMP panic In-Reply-To: <199709042322.RAA11106@Ilsa.StevesCafe.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Yep, this ends up being true. I increased the load by mailing larger messages, and it crashed with nearly identical panic. If somebody can give me what they want done at ddb>, I'm more than happy to send in the output. For now, I'm going to boot UNI and see what happens. On Thu, 4 Sep 1997, Steve Passe wrote: > Hi, > > > Recent kernel changes (maybe the tcp stuff garrett did relating to > > software interrupts? (I have a DPT controller), seems to have fixed it, > > The same load that was crashing it yesterday after supping and hbuilding > > and booting new kernel is running fine. > > unfortunately not, I have a current sys tree from about 2 hours ago and I can > get the panic by loading the daemon screen saver. it probably just moved > things around in a way that hides the problem for you... > -- > Steve Passe | powered by > smp@csn.net | Symmetric MultiProcessor FreeBSD > > From owner-freebsd-smp Thu Sep 4 17:19:46 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id RAA03180 for smp-outgoing; Thu, 4 Sep 1997 17:19:46 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id RAA03175 for ; Thu, 4 Sep 1997 17:19:39 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id SAA11354; Thu, 4 Sep 1997 18:19:24 -0600 (MDT) Message-Id: <199709050019.SAA11354@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Jaye Mathisen cc: "John S. Dyson" , smp@FreeBSD.ORG Subject: Re: 3.0/SMP panic In-reply-to: Your message of "Thu, 04 Sep 1997 17:03:20 PDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 04 Sep 1997 18:19:24 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, I've undefined almost everything from smptests.h, effectively reverting to one giant lock on everything, including FAST_INTR()s. The realignment of INT priorities are out, the COM, clock and INTR locks are out. The skipping of ipending checks in FAST_INTR() is out. (Note that to undefine all these I had to tweek a couple of source files.) I am still getting a panic from the daemon saver, so I'm fairly certain its not from my recent changes (but never say never...) So I guess I'm going after the saver itself next... -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Thu Sep 4 20:37:39 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id UAA13265 for smp-outgoing; Thu, 4 Sep 1997 20:37:39 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id UAA13260 for ; Thu, 4 Sep 1997 20:37:36 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.5/8.6.9) id NAA17249; Fri, 5 Sep 1997 13:29:36 +1000 Date: Fri, 5 Sep 1997 13:29:36 +1000 From: Bruce Evans Message-Id: <199709050329.NAA17249@godzilla.zeta.org.au> To: mrcpu@cdsnet.net, smp@csn.net Subject: Re: 3.0/SMP panic Cc: smp@FreeBSD.ORG, toor@dyson.iquest.net Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >I am still getting a panic from the daemon saver, so I'm fairly certain >its not from my recent changes (but never say never...) > >So I guess I'm going after the saver itself next... In general you have to worry about LKMs calling disable_intr() or other SMP-ifdefed inlined functions or macros. The screen savers don't call disable_intr(). The joy module calls it. Bruce From owner-freebsd-smp Fri Sep 5 12:10:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id MAA26023 for smp-outgoing; Fri, 5 Sep 1997 12:10:04 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id MAA25960; Fri, 5 Sep 1997 12:09:42 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id NAA14813; Fri, 5 Sep 1997 13:09:37 -0600 (MDT) Message-Id: <199709051909.NAA14813@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: Jaye Mathisen cc: scsi@FreeBSD.ORG, smp@FreeBSD.ORG, current@FreeBSD.ORG Subject: Re: Scary DPT problem. In-reply-to: Your message of "Fri, 05 Sep 1997 10:18:34 PDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 05 Sep 1997 13:09:37 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > Using DPT-1.2.4 with 3.0/SMP. > > I was running a test of adding 25000 accounts and removing them, and > making sure everything was hunky-dory, when the dreaded: > > "DPT: Undocumented Error" > db> > > Occurs. > > OK, Happened in the past. Hit the reset switch: "No Operating System". > Power-Cycle "No Operating System". Swear, curse, cry, yell, shake. > > Power-cycle it, and let it sit. It finally reboots, but the passwd file > .db's are completely hosed. and /etc/group is a goner. > > Makes me nervous to deploy this out in the sticks... I'm hoping it's just > an SMP or 3.0-current problem, as opposed to some deep dark nasty DPT > issue. SMP-current should definitely NOT be used for anything critical right now! We KNOW we have a show-stopper in there... It probably affects ALL SMP systems, just bites some earlier than others. On the bright side, we are closing in on it, and hopefully will have a resolution by later today (no promises). -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-smp Fri Sep 5 12:59:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id MAA28270 for smp-outgoing; Fri, 5 Sep 1997 12:59:04 -0700 (PDT) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id MAA28240; Fri, 5 Sep 1997 12:58:53 -0700 (PDT) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.6/8.8.6) with SMTP id MAA09927; Fri, 5 Sep 1997 12:58:47 -0700 (PDT) Date: Fri, 5 Sep 1997 12:58:45 -0700 (PDT) From: Jaye Mathisen To: Steve Passe cc: scsi@FreeBSD.ORG, smp@FreeBSD.ORG, current@FreeBSD.ORG Subject: Re: Scary DPT problem. In-Reply-To: <199709051909.NAA14813@Ilsa.StevesCafe.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk My apologies. I'm well aware of the purpose of -current vs -stable. I was more concerned with the fact that it appears to be possible to get the DPT controller so scrambled up that a simple reboot doesn't reset it. Not anything with SMP/3.0... FWIW, I can still crash SMP/3.0 if I bump up InetLoad high enough, but I still can't crash a 3.0-uniprocessor kernel with no MAXMEM... Haven't supped since yesterday, given the headaches of applying the DPT patches all the time to the sys tree. I don't see this so much as a 3.0 problem, as a DPT issue, but I could be off base. Wouldn't be the first time... :0 On Fri, 5 Sep 1997, Steve Passe wrote: > Hi, > > > Using DPT-1.2.4 with 3.0/SMP. > > > > I was running a test of adding 25000 accounts and removing them, and > > making sure everything was hunky-dory, when the dreaded: > > > > "DPT: Undocumented Error" > > db> > > > > Occurs. > > > > OK, Happened in the past. Hit the reset switch: "No Operating System". > > Power-Cycle "No Operating System". Swear, curse, cry, yell, shake. > > > > Power-cycle it, and let it sit. It finally reboots, but the passwd file > > .db's are completely hosed. and /etc/group is a goner. > > > > Makes me nervous to deploy this out in the sticks... I'm hoping it's just > > an SMP or 3.0-current problem, as opposed to some deep dark nasty DPT > > issue. > > SMP-current should definitely NOT be used for anything critical right now! > We KNOW we have a show-stopper in there... It probably affects ALL SMP > systems, just bites some earlier than others. > > On the bright side, we are closing in on it, and hopefully will have a > resolution by later today (no promises). > -- > Steve Passe | powered by > smp@csn.net | Symmetric MultiProcessor FreeBSD > > From owner-freebsd-smp Fri Sep 5 14:31:58 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA02462 for smp-outgoing; Fri, 5 Sep 1997 14:31:58 -0700 (PDT) Received: from mx.serv.net (mx.serv.net [205.153.153.234]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA02456 for ; Fri, 5 Sep 1997 14:31:55 -0700 (PDT) Received: from aegis (mercury.aegis1.com [205.153.153.45]) by mx.serv.net (8.8.5/8.8.5) with SMTP id OAA24653 for ; Fri, 5 Sep 1997 14:31:59 -0700 (PDT) Received: by localhost with Microsoft MAPI; Fri, 5 Sep 1997 14:34:41 -0700 Message-ID: <01BCBA08.D850F420.adam@aegis1.com> From: "Adam J. Bartels" To: "smp@FreeBSD.ORG" Subject: unsubscribe smp adam@aegis1.com Date: Fri, 5 Sep 1997 14:22:05 -0700 X-Mailer: Microsoft Internet E-mail/MAPI - 8.0.0.4211 Encoding: 1 TEXT Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk unsubscribe smp adam@aegis1.com From owner-freebsd-smp Fri Sep 5 22:05:10 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA22564 for smp-outgoing; Fri, 5 Sep 1997 22:05:10 -0700 (PDT) Received: from sendero-ppp.i-connect.net (sendero-ppp.i-Connect.Net [206.190.143.100]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id WAA22503 for ; Fri, 5 Sep 1997 22:04:53 -0700 (PDT) Received: (qmail 16777 invoked by uid 1000); 6 Sep 1997 05:04:59 -0000 Message-ID: X-Mailer: XFMail 1.2-alpha [p0] on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199709051909.NAA14813@Ilsa.StevesCafe.com> Date: Fri, 05 Sep 1997 22:04:59 -0700 (PDT) Organization: Atlas Telecom From: Simon Shapiro To: Steve Passe Subject: Re: Scary DPT problem. Cc: current@FreeBSD.ORG, smp@FreeBSD.ORG, scsi@FreeBSD.ORG, Jaye Mathisen Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi Steve Passe; On 05-Sep-97 you wrote: ... > > "DPT: Undocumented Error" This is when we get an error indication from the DPT controller, but have no clue what it is. ... > SMP-current should definitely NOT be used for anything critical right > now! To add to Steve's message, I will quote the copyright notice on 1.2.4 DPT driver: .... * This is a proprietary, unpublished source code. No publishing, copying, * distribution or use permission is granted to anyone. * * If you want to use this product in any way, please contact the author by * sending email to shimon@i-connect.net .... The 1.2.4 driver (as you can clearly see is NOT in the public domain! Nor is it something you should grab, use without permission on an experimental version of the O/S, mess up something, complain about and recive much sympathy. I posted it on my personal machine as a means for experimenters who are generous and gracious enough to assist me in debugging it. --- Sincerely Yours, (Sent on 05-Sep-97, 21:49:22 by XF-Mail) Simon Shapiro Atlas Telecom Senior Architect 14355 SW Allen Blvd., Suite 130 Beaverton OR 97005 Shimon@i-Connect.Net Voice: 503.643.5559, Emergency: 503.799.2313 From owner-freebsd-smp Fri Sep 5 22:05:02 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA22537 for smp-outgoing; Fri, 5 Sep 1997 22:05:02 -0700 (PDT) Received: from sendero-ppp.i-connect.net (sendero-ppp.i-Connect.Net [206.190.143.100]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id WAA22494 for ; Fri, 5 Sep 1997 22:04:50 -0700 (PDT) Received: (qmail 16779 invoked by uid 1000); 6 Sep 1997 05:04:59 -0000 Message-ID: X-Mailer: XFMail 1.2-alpha [p0] on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: Date: Fri, 05 Sep 1997 22:04:59 -0700 (PDT) Organization: Atlas Telecom From: Simon Shapiro To: Jaye Mathisen Subject: Re: Scary DPT problem. Cc: current@FreeBSD.ORG, smp@FreeBSD.ORG, scsi@FreeBSD.ORG, Steve Passe Sender: owner-freebsd-smp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi Jaye Mathisen; On 05-Sep-97 you wrote: > > My apologies. I'm well aware of the purpose of -current vs -stable. I > was more concerned with the fact that it appears to be possible to get > the > DPT controller so scrambled up that a simple reboot doesn't reset it. It (the DPT controller) is a computer with a precise and specific sequencer implemented in the EATA protocol. If the O/S is doing something strange, it may decide to stop playing. Besides, you have no real proof what hung up. I am not going to consider this a DPT driver bug until I see more data. > I don't see this so much as a 3.0 problem, as a DPT issue, but I could > be > off base. Wouldn't be the first time... :0 You are off base in this particular case. --- Sincerely Yours, (Sent on 05-Sep-97, 21:55:33 by XF-Mail) Simon Shapiro Atlas Telecom Senior Architect 14355 SW Allen Blvd., Suite 130 Beaverton OR 97005 Shimon@i-Connect.Net Voice: 503.643.5559, Emergency: 503.799.2313 From owner-freebsd-smp Sat Sep 6 13:45:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id NAA00968 for smp-outgoing; Sat, 6 Sep 1997 13:45:55 -0700 (PDT) Received: from tahoe.cinenet.net (root@ns1.cinenet.net [198.147.76.65]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id NAA00963 for ; Sat, 6 Sep 1997 13:45:52 -0700 (PDT) Received: from hollywood.cinenet.net (hollywood.cinenet.net [198.147.76.75]) by tahoe.cinenet.net (8.8.6/8.8.6) with ESMTP id NAA16733 for ; Sat, 6 Sep 1997 13:45:50 -0700 (PDT) From: Bob Mercier Received: (mercier@localhost) by hollywood.cinenet.net (8.7.3/8.6.9) id NAA28395 for smp@freebsd.org; Sat, 6 Sep 1997 13:46:10 -0700 (PDT) Date: Sat, 6 Sep 1997 13:46:10 -0700 (PDT) Message-Id: <199709062046.NAA28395@hollywood.cinenet.net> To: smp@freebsd.org Subject: subscribe Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk subscribe From owner-freebsd-smp Sat Sep 6 19:44:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id TAA21039 for smp-outgoing; Sat, 6 Sep 1997 19:44:13 -0700 (PDT) Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id TAA20996; Sat, 6 Sep 1997 19:44:01 -0700 (PDT) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.8.7/8.8.5) with ESMTP id UAA23614; Sat, 6 Sep 1997 20:43:59 -0600 (MDT) Message-Id: <199709070243.UAA23614@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0gamma 1/27/96 From: Steve Passe To: smp@freebsd.org cc: current@freebsd.org Subject: FIXED: Fatal trap 12 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 06 Sep 1997 20:43:59 -0600 Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, John found and fixed the "Fatal trap 12" bug in SMP. It fixes the problem for both his and my test cases. The fix has been committed to freefall. For the impatient behind slow mirrors, the patch: ------------------------------------ cut ------------------------------------ *** pmap.c.orig 1997/09/07 01:55:48 --- pmap.c 1997/09/07 01:55:57 *************** *** 1464,1470 **** pdir_pde(PTD, kernel_vm_end) = (pd_entry_t) (VM_PAGE_TO_PHYS(nkp g) | PG_V | PG_RW | pgeflag); #ifdef SMP ! for (i = 0; i < mp_naps; i++) { if (IdlePTDS[i]) pdir_pde(IdlePTDS[i], kernel_vm_end) = (pd_entry _t) (VM_PAGE_TO_PHYS(nkpg) | PG_V | PG_RW | pgeflag); } --- 1464,1470 ---- pdir_pde(PTD, kernel_vm_end) = (pd_entry_t) (VM_PAGE_TO_PHYS(nkp g) | PG_V | PG_RW | pgeflag); #ifdef SMP ! for (i = 0; i < mp_ncpus; i++) { if (IdlePTDS[i]) pdir_pde(IdlePTDS[i], kernel_vm_end) = (pd_entry _t) (VM_PAGE_TO_PHYS(nkpg) | PG_V | PG_RW | pgeflag); } ------------------------------------ cut ------------------------------------ -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD