From owner-freebsd-smp Tue Jun 19 6:16:17 2001 Delivered-To: freebsd-smp@freebsd.org Received: from smtp-mhd.dmz.hitachi-eu.com (gatekeeper.hitachi-eu.com [194.131.190.1]) by hub.freebsd.org (Postfix) with ESMTP id AB6CE37B403 for ; Tue, 19 Jun 2001 06:16:12 -0700 (PDT) (envelope-from eva.coleman-daniels@hitachi-eu.com) Received: from mimesweeper2.hitachi-eu.com ([193.39.225.234]) by smtp-mhd.dmz.hitachi-eu.com (Netscape Messaging Server 4.15) with ESMTP id GF6I3300.6RU for ; Tue, 19 Jun 2001 14:13:51 +0100 Received: from mail-mhd.hitachi-eu.com (unverified) by mimesweeper2.hitachi-eu.com (Content Technologies SMTPRS 4.1.5) with ESMTP id for ; Tue, 19 Jun 2001 14:14:46 +0100 Received: from hitachi-eu.com ([192.168.19.169]) by mail-mhd.hitachi-eu.com (Netscape Messaging Server 4.15) with ESMTP id GF6I2T00.S4Q for ; Tue, 19 Jun 2001 14:13:41 +0100 Message-ID: <3B2F5002.5B7C91BC@hitachi-eu.com> Date: Tue, 19 Jun 2001 14:13:38 +0100 From: "Eva Coleman-Daniels" Organization: Hitachi Europe Ltd X-Mailer: Mozilla 4.77 [en]C-CCK-MCD NECCK (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: freebsd-smp@FreeBSD.ORG Subject: Re: Soyo Motherboard Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org When I installed a 20G hardrive in my pc it only recognises 504mb. My motherboard is a Soyo motherboard. I am not sure where to look for information about the type. However, when I started the PC, for the BIOS info I had the following info: Rev. UA2.2 (s/I/m) I also saw the following info on the board Intel PCI Set S882371FB L6123006 Intel 94 Pls help, as I would like to flush the bioa -- Eva Coleman-Daniels Hitachi Europe Ltd (Support Engineer) Tel: (44) 020 7970 2726 Mobile: 07796996177 Fax: (44) 020 790 2799 ********************************************************************* E-mail Confidentiality Notice and Disclaimer This email and any files transmitted with it are confidential and are intended solely for the use of the individual or entity to which they are addressed. Access to this e-mail by anyone else is unauthorised. If you are not the intended recipient, any disclosure, copying, distribution or any action taken or omitted to be taken in reliance on it, is prohibited. E-mail messages are not necessarily secure. Hitachi does not accept responsibility for any changes made to this message after it was sent. Please note that Hitachi checks outgoing e-mail messages for the presence of computer viruses. ********************************************************************* To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Jun 20 11:23:25 2001 Delivered-To: freebsd-smp@freebsd.org Received: from carp.aciri.org (carp.aciri.org [192.150.187.71]) by hub.freebsd.org (Postfix) with ESMTP id 3A54E37B401 for ; Wed, 20 Jun 2001 11:23:23 -0700 (PDT) (envelope-from bkarp@carp.aciri.org) Received: from carp.aciri.org (localhost [127.0.0.1]) by carp.aciri.org (8.11.3/8.11.1) with ESMTP id f5KINN321282 for ; Wed, 20 Jun 2001 11:23:23 -0700 (PDT) (envelope-from bkarp@carp.aciri.org) Message-Id: <200106201823.f5KINN321282@carp.aciri.org> X-Mailer: exmh version 2.1.1 10/15/1999 From: Brad Karp To: smp@freebsd.org Subject: clock speedup on SMP boxes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 20 Jun 2001 11:23:23 -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org There were several responses to my initial note that the time of day clock in FreeBSD 4.3-RELEASE runs faster than real time (gaining up to an hour every day, where the amount of clock speedup is in proportion to the load on the CPUs) on my SMP boxes. In these responses, others mentioned they see similar behavior on other hardware. I see this behavior on Intel STL2 and Tyan S2567 motherboards with dual 1 GHz P-IIIs installed. To wit, the archives: http://docs.freebsd.org/mail/archive/2001/freebsd-smp/20010617.freebsd-smp.html Is there any SMP developer I should directly notify (though I presume he already reads -smp), in the interest of having this fixed for all SMP users? Or is it low priority, compared with other SMP development work underway? Or is there reason to believe that -CURRENT would behave better? Thanks, -Brad, bkarp@icsi.berkeley.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Jun 20 11:56:24 2001 Delivered-To: freebsd-smp@freebsd.org Received: from meow.osd.bsdi.com (meow.osd.bsdi.com [204.216.28.88]) by hub.freebsd.org (Postfix) with ESMTP id 790DA37B401 for ; Wed, 20 Jun 2001 11:56:22 -0700 (PDT) (envelope-from jhb@FreeBSD.org) Received: from laptop.baldwin.cx (john@jhb-laptop.osd.bsdi.com [204.216.28.241]) by meow.osd.bsdi.com (8.11.3/8.11.2) with ESMTP id f5KIthI31218; Wed, 20 Jun 2001 11:55:43 -0700 (PDT) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.4.0 on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <200106201823.f5KINN321282@carp.aciri.org> Date: Wed, 20 Jun 2001 12:01:30 -0700 (PDT) From: John Baldwin To: Brad Karp Subject: RE: clock speedup on SMP boxes Cc: smp@FreeBSD.org Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On 20-Jun-01 Brad Karp wrote: > Is there any SMP developer I should directly notify (though I presume he > already reads -smp), in the interest of having this fixed for all SMP users? > Or is it low priority, compared with other SMP development work underway? Most of the SMP developers already read -smp. I, personally, cannot think of a reason off the top of my head why the clock would be sped up under a loaded SMP system. > Or is there reason to believe that -CURRENT would behave better? s/better/different/ It's better in some ways, but not in others right now. > Thanks, > -Brad, bkarp@icsi.berkeley.edu -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.baldwin.cx/~john/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Jun 20 12:40:58 2001 Delivered-To: freebsd-smp@freebsd.org Received: from critter.freebsd.dk (critter.freebsd.dk [212.242.86.163]) by hub.freebsd.org (Postfix) with ESMTP id 9F80837B403 for ; Wed, 20 Jun 2001 12:40:53 -0700 (PDT) (envelope-from phk@critter.freebsd.dk) Received: from critter (localhost [127.0.0.1]) by critter.freebsd.dk (8.11.3/8.11.3) with ESMTP id f5KJem607316; Wed, 20 Jun 2001 21:40:48 +0200 (CEST) (envelope-from phk@critter.freebsd.dk) To: Brad Karp Cc: smp@FreeBSD.ORG Subject: Re: clock speedup on SMP boxes In-Reply-To: Your message of "Wed, 20 Jun 2001 11:23:23 PDT." <200106201823.f5KINN321282@carp.aciri.org> Date: Wed, 20 Jun 2001 21:40:48 +0200 Message-ID: <7314.993066048@critter> From: Poul-Henning Kamp Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org In message <200106201823.f5KINN321282@carp.aciri.org>, Brad Karp writes: >There were several responses to my initial note that the time of day clock in >FreeBSD 4.3-RELEASE runs faster than real time (gaining up to an hour every >day, where the amount of clock speedup is in proportion to the load on the >CPUs) on my SMP boxes. In these responses, others mentioned they see similar >behavior on other hardware. I see this behavior on Intel STL2 and Tyan S2567 >motherboards with dual 1 GHz P-IIIs installed. I'm the author of our timekeeping ("timecounter") code, I heard about this problem first time today. Unfortunately part of my lab is already packed down so I can't reproduce this locally right now. I don't have a good idea right now what causes this, but I have a number of ideas I would like people to try out: 0. First off, kill ntpd/ntpdate-from-cron. 1. Try to quanitify the speedup at 100% CPU load with no interrupts. Basically run one copy of for each CPU in the system: main() { for(;;); } 2. Try to quanitify the speedup at 1/NCPU load with no interrupts. Run just one copy of the above. 3. Increase kernel HZ to 500 and rerun #2. 4. Decrease kernel HZ to 50 and rerun #2. Please include in your report the contents of /var/run/dmesg.boot please email your data to timec@phk.freebsd.dk I'll try to see if I can find out what's going on. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Jun 20 15: 2:30 2001 Delivered-To: freebsd-smp@freebsd.org Received: from dan.emsphone.com (dan.emsphone.com [199.67.51.101]) by hub.freebsd.org (Postfix) with ESMTP id AF43D37B401; Wed, 20 Jun 2001 15:02:26 -0700 (PDT) (envelope-from dan@dan.emsphone.com) Received: (from dan@localhost) by dan.emsphone.com (8.11.3/8.11.3) id f5KM2Ls22999; Wed, 20 Jun 2001 17:02:21 -0500 (CDT) (envelope-from dan) Date: Wed, 20 Jun 2001 17:02:21 -0500 From: Dan Nelson To: John Baldwin Cc: Brad Karp , smp@FreeBSD.ORG Subject: Re: clock speedup on SMP boxes Message-ID: <20010620170220.A19968@dan.emsphone.com> References: <200106201823.f5KINN321282@carp.aciri.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.3.19i X-OS: FreeBSD 5.0-CURRENT Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org In the last episode (Jun 20), John Baldwin said: > On 20-Jun-01 Brad Karp wrote: > > Is there any SMP developer I should directly notify (though I > > presume he already reads -smp), in the interest of having this > > fixed for all SMP users? Or is it low priority, compared with other > > SMP development work underway? > > Most of the SMP developers already read -smp. I, personally, cannot > think of a reason off the top of my head why the clock would be sped > up under a loaded SMP system. > > > Or is there reason to believe that -CURRENT would behave better? I don't have any clock skew problems on two Dell PowerEdge 1300 2-cpu boxes running 4.3, nor on a SuperMicro 2-cpu box running -current. All 3 occasionally run at 100% cpu for hours at a time. -- Dan Nelson dnelson@emsphone.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Jun 20 15:17:59 2001 Delivered-To: freebsd-smp@freebsd.org Received: from carp.aciri.org (carp.aciri.org [192.150.187.71]) by hub.freebsd.org (Postfix) with ESMTP id 62B4737B408 for ; Wed, 20 Jun 2001 15:17:56 -0700 (PDT) (envelope-from bkarp@carp.aciri.org) Received: from carp.aciri.org (localhost [127.0.0.1]) by carp.aciri.org (8.11.3/8.11.1) with ESMTP id f5KMHs323802; Wed, 20 Jun 2001 15:17:54 -0700 (PDT) (envelope-from bkarp@carp.aciri.org) Message-Id: <200106202217.f5KMHs323802@carp.aciri.org> X-Mailer: exmh version 2.1.1 10/15/1999 From: Brad Karp To: Dan Nelson Cc: smp@freebsd.org Subject: Re: clock speedup on SMP boxes In-reply-to: Your message of "Wed, 20 Jun 2001 17:02:21 CDT." <20010620170220.A19968@dan.emsphone.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 20 Jun 2001 15:17:54 -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org > I don't have any clock skew problems on two Dell PowerEdge 1300 2-cpu > boxes running 4.3, nor on a SuperMicro 2-cpu box running -current. > All 3 occasionally run at 100% cpu for hours at a time. We have several Dell multi-processor boxes here that also do not suffer from the clock speedup problem. I didn't mean to imply that *all* SMP motherboards exhibit this behavior; only that a set of several do (exemplified by my reports regarding the Intel STL2 and Tyan S2567, and others' reports of this behavior on other motherboards). It would appear that the delivery or processing of timecounter interrupts is not correct on some SMP motherboards, but correct on others. -Brad, bkarp@icsi.berkeley.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jun 21 5:35:15 2001 Delivered-To: freebsd-smp@freebsd.org Received: from mail.imp.ch (mail.imp.ch [157.161.1.2]) by hub.freebsd.org (Postfix) with ESMTP id CC7B137B401 for ; Thu, 21 Jun 2001 05:35:10 -0700 (PDT) (envelope-from mb@imp.ch) Received: from levais.imp.ch (levais.imp.ch [157.161.4.66]) by mail.imp.ch (8.11.1/8.11.1) with ESMTP id f5LCZAl24383 for ; Thu, 21 Jun 2001 14:35:10 +0200 (CEST) (envelope-from Martin.Blapp@imp.ch) Date: Thu, 21 Jun 2001 14:39:09 +0200 (CEST) From: Martin Blapp To: Subject: RE: clock speedup on SMP boxes Message-ID: <20010621143000.I766-100000@levais.imp.ch> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Hi, We have seen this timedrifting here too on a 4.3 RELEASE. We have some busy SMP mysql servers, and from time to time a mysql prozess hangs get stuck in poll() and gettimeofday(). With a normal for loop, or a busy system we do not see a noticable time drifting, but if we start this (below attached) little program here, we see a severe time drifting, the clock runs much faster ! Our servers here are big SMP servers from Siemens. Here is some dmesg output: Timecounter "i8254" frequency 1193182 Hz CPU: Pentium III/Pentium III Xeon/Celeron (933.37-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x686 Stepping = 6 Features=0x383fbff real memory = 1073721344 (1048556K bytes) avail memory = 1040846848 (1016452K bytes) Changing APIC ID for IO APIC #1 from 3 to 1 in MP table Changing APIC ID for IO APIC #1 from 3 to 1 on chip Programming 16 pins in IOAPIC #0 IOAPIC #0 intpin 2 -> irq 0 Programming 16 pins in IOAPIC #1 FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 3, version: 0x00040011, at 0xfee00000 cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 io0 (APIC): apic id: 2, version: 0x000f0011, at 0xfec00000 io1 (APIC): apic id: 1, version: 0x000f0011, at 0xfec01000 IOAPIC #1 intpin 4 -> irq 2 IOAPIC #1 intpin 2 -> irq 10 APIC_IO: Testing 8254 interrupt delivery APIC_IO: Broken MP table detected: 8254 is not connected to IOAPIC #0 intpin 2 APIC_IO: routing 8254 via 8259 and IOAPIC #0 intpin 0 And here the test programm which produces the time drifting. #include #include #include #include #include #define DATA "/dev/zero" int main(void) { int fd; struct pollfd pfd; fd = open(DATA, O_RDONLY); if(fd < 0) { err(1, "open()"); } for(;;) { int nfds; struct timeval tv; struct timezone tz; pfd.fd = fd; pfd.events = POLLRDNORM; pfd.revents = 0; nfds = poll(&pfd, 1, 0); gettimeofday(&tv, &tz); } return (0); } Martin Blapp, mb@imp.ch ------------------------------------------------ Improware AG, UNIX solution and service provider Zurlindenstrasse 29, 4133 Pratteln, Switzerland Phone: +41 79 370 26 05, Fax: +41 61 826 93 01 ------------------------------------------------ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jun 21 5:47:46 2001 Delivered-To: freebsd-smp@freebsd.org Received: from mail.imp.ch (mail.imp.ch [157.161.1.2]) by hub.freebsd.org (Postfix) with ESMTP id E045837B401; Thu, 21 Jun 2001 05:47:40 -0700 (PDT) (envelope-from mb@imp.ch) Received: from levais.imp.ch (levais.imp.ch [157.161.4.66]) by mail.imp.ch (8.11.1/8.11.1) with ESMTP id f5LCldl25676; Thu, 21 Jun 2001 14:47:39 +0200 (CEST) (envelope-from Martin.Blapp@imp.ch) Date: Thu, 21 Jun 2001 14:51:38 +0200 (CEST) From: Martin Blapp To: Cc: Brad Karp , Subject: RE: clock speedup on SMP boxes In-Reply-To: <20010621143000.I766-100000@levais.imp.ch> Message-ID: <20010621144914.P766-100000@levais.imp.ch> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Hi, Just gettimeofday() produces 8sec time drifting now. No need to use poll() in our little programm I sent previously. There is no time drifting if we used a 100% load programm with just poll(). Very strange. Do you have some idea ? Martin Martin Blapp, mb@imp.ch ------------------------------------------------ Improware AG, UNIX solution and service provider Zurlindenstrasse 29, 4133 Pratteln, Switzerland Phone: +41 79 370 26 05, Fax: +41 61 826 93 01 ------------------------------------------------ > #include > #include > #include > #include > #include > > #define DATA "/dev/zero" > > int > main(void) > { > int fd; > struct pollfd pfd; > > fd = open(DATA, O_RDONLY); > if(fd < 0) { > err(1, "open()"); > } > for(;;) { > int nfds; > struct timeval tv; > struct timezone tz; > > pfd.fd = fd; > pfd.events = POLLRDNORM; > pfd.revents = 0; > > nfds = poll(&pfd, 1, 0); > gettimeofday(&tv, &tz); > } > return (0); > } To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jun 21 6:51: 1 2001 Delivered-To: freebsd-smp@freebsd.org Received: from mailman.zeta.org.au (mailman.zeta.org.au [203.26.10.16]) by hub.freebsd.org (Postfix) with ESMTP id 99E8C37B403; Thu, 21 Jun 2001 06:50:58 -0700 (PDT) (envelope-from bde@zeta.org.au) Received: from bde.zeta.org.au (bde.zeta.org.au [203.2.228.102]) by mailman.zeta.org.au (8.9.3/8.8.7) with ESMTP id XAA02953; Thu, 21 Jun 2001 23:50:54 +1000 Date: Thu, 21 Jun 2001 23:49:04 +1000 (EST) From: Bruce Evans X-Sender: bde@besplex.bde.org To: Martin Blapp Cc: smp@FreeBSD.ORG, Brad Karp , phk@FreeBSD.ORG Subject: RE: clock speedup on SMP boxes In-Reply-To: <20010621144914.P766-100000@levais.imp.ch> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Thu, 21 Jun 2001, Martin Blapp wrote: > Just gettimeofday() produces 8sec time drifting now. No need > to use poll() in our little programm I sent previously. > > There is no time drifting if we used a 100% load programm with > just poll(). > > Very strange. Do you have some idea ? From clock.c in -current: | #ifdef APIC_IO | #define lapic_irr1 ((volatile u_int *)&lapic)[0x210 / 4] /* XXX XXX */ | /* XXX this assumes that apic_8254_intr is < 24. */ | (lapic_irr1 & (1 << apic_8254_intr)))) | #else | (inb(IO_ICU1) & 1))) | #endif Maybe apic_8254_intr is not < 24. I think the second XXX comment has rotted in -current, but it still applies in RELENG_4. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jun 21 8: 8:21 2001 Delivered-To: freebsd-smp@freebsd.org Received: from mail.imp.ch (mail.imp.ch [157.161.1.2]) by hub.freebsd.org (Postfix) with ESMTP id 0846B37B401; Thu, 21 Jun 2001 08:08:17 -0700 (PDT) (envelope-from mb@imp.ch) Received: from levais.imp.ch (levais.imp.ch [157.161.4.66]) by mail.imp.ch (8.11.1/8.11.1) with ESMTP id f5LF88039619; Thu, 21 Jun 2001 17:08:08 +0200 (CEST) (envelope-from Martin.Blapp@imp.ch) Date: Thu, 21 Jun 2001 17:12:08 +0200 (CEST) From: Martin Blapp To: Bruce Evans Cc: , Brad Karp , Subject: RE: clock speedup on SMP boxes In-Reply-To: Message-ID: <20010621170959.V766-100000@levais.imp.ch> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Hi, > Just gettimeofday() produces 8sec time drifting now. No need > to use poll() in our little programm I sent previously. And this is a lot more strange ! If we start the gettimeofday() loop once (on one CPU) I get the timedrift with 8 seconds per minute. If we start it twice so it runs on both CPU's, there is no significant timedrift !!! Something here is cracy. Martin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jun 21 8:23:34 2001 Delivered-To: freebsd-smp@freebsd.org Received: from relay.tecc.co.uk (luggage.tecc.co.uk [193.128.6.129]) by hub.freebsd.org (Postfix) with SMTP id BD97137B403; Thu, 21 Jun 2001 08:23:18 -0700 (PDT) (envelope-from andy@tecc.co.uk) Received: from southampton [195.217.37.155] by relay.tecc.co.uk with smtp (Exim 1.70 #1) id 15D6If-0002yu-00; Thu, 21 Jun 2001 16:23:17 +0100 From: "Andy" To: Cc: , Subject: RE: shutdown not completing Date: Thu, 21 Jun 2001 16:23:17 +0100 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) In-Reply-To: Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org cross posting this to -smp and -hackers as it seems to be a problem there..... for those who are new to the problem, in -stable we have reports of "shutdown now" hanging rather than dropping to single user mode. I have a dual PIII machine. At 4.2-RELEASE I had no problems at all. However, two/three days ago I went to -STABLE. I now have the "shutdown now" problem. However, if I boot GENERIC I have no problem with shutdown now. So, as an experiment I made a new kernel based on my SMP kernel. The only change I made was to drop the two lines which make it an SMP kernel. On doing "shutdown now" with this kernel I go to single user mode. So, the "shutdown now" problem appears to be connected with shutting down processor #1 ?? Anyone shed light on this? Regards Andy To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jun 21 8:50: 6 2001 Delivered-To: freebsd-smp@freebsd.org Received: from mikea.ath.cx (okc-65-26-223-53.mmcable.com [65.26.223.53]) by hub.freebsd.org (Postfix) with ESMTP id E0B6F37B401; Thu, 21 Jun 2001 08:49:57 -0700 (PDT) (envelope-from mikea@mikea.ath.cx) Received: (from mikea@localhost) by mikea.ath.cx (8.11.4/8.11.1) id f5LFnvD66024; Thu, 21 Jun 2001 10:49:57 -0500 (CDT) (envelope-from mikea) Date: Thu, 21 Jun 2001 10:49:57 -0500 From: mikea Cc: freebsd-stable@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: shutdown not completing Message-ID: <20010621104957.C65743@mikea.ath.cx> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.4i In-Reply-To: ; from andy@tecc.co.uk on Thu, Jun 21, 2001 at 04:23:17PM +0100 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Thu, Jun 21, 2001 at 04:23:17PM +0100, Andy wrote: > cross posting this to -smp and -hackers as it > seems to be a problem there..... > > for those who are new to the problem, in -stable > we have reports of "shutdown now" hanging rather > than dropping to single user mode. > > I have a dual PIII machine. At 4.2-RELEASE I had > no problems at all. However, two/three days ago > I went to -STABLE. I now have the "shutdown now" > problem. > > However, if I boot GENERIC I have no problem with > shutdown now. So, as an experiment I made a new > kernel based on my SMP kernel. The only change I > made was to drop the two lines which make it an > SMP kernel. > > On doing "shutdown now" with this kernel I go > to single user mode. > > So, the "shutdown now" problem appears to be > connected with shutting down processor #1 ?? > > Anyone shed light on this? Another data point: I'm running 4.3-stable, cvsup'd to the 20010610 code, on an MP kernel (MSI 694D-AR motherboard, 2xPiii-866). I have not experienced any problem in shutdown, and have been fiddling with things a lot recently. Some days I've done two or three shutdowns to single-user mode and multiple shutdowns to the completely-down "hit-any-key" state. -- Mike Andrews mikea@mikea.ath.cx Tired old sysadmin since 1964 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jun 21 9:54:48 2001 Delivered-To: freebsd-smp@freebsd.org Received: from relay.tecc.co.uk (luggage.tecc.co.uk [193.128.6.129]) by hub.freebsd.org (Postfix) with SMTP id BC0A537B406; Thu, 21 Jun 2001 09:54:39 -0700 (PDT) (envelope-from andy@tecc.co.uk) Received: from southampton [195.217.37.155] by relay.tecc.co.uk with smtp (Exim 1.70 #1) id 15D7j4-0003t1-00; Thu, 21 Jun 2001 17:54:38 +0100 From: "Andy" To: Cc: , Subject: RE: shutdown not completing (more info) Date: Thu, 21 Jun 2001 17:54:38 +0100 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) In-Reply-To: Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Install: 4.2-RELEASE from CDRom, GENERIC kernel "shutdown now" works fine. 4.3-RELEASE from CDRom, GENERIC kernel "shutdown now" fails, hangs machine. It seems I was outa touch with the SMP idea. The 4.3-RELEASE generic kernel (without smp) causes the problem so I won't cross post this to -smp anymore. But there does appear to be a problem. The above installs were done on the same hardware. I'll start trying to get closer to it but for now, believe us minority when we say "shutdown now" doesn't work but just hangs the system. fwiw, all these *do* work....... "shutdown -h now" "shutdown -r now" "reboot" "halt" To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jun 21 14:42: 7 2001 Delivered-To: freebsd-smp@freebsd.org Received: from beastie.saturn-tech.com (beastie.saturn-tech.com [207.229.19.136]) by hub.freebsd.org (Postfix) with ESMTP id D755237B401 for ; Thu, 21 Jun 2001 14:42:03 -0700 (PDT) (envelope-from drussell@saturn-tech.com) Received: from localhost (drussell@localhost) by beastie.saturn-tech.com (8.11.1/8.11.0) with ESMTP id f5LLkCs43608 for ; Thu, 21 Jun 2001 15:46:12 -0600 (MDT) (envelope-from drussell@saturn-tech.com) X-Authentication-Warning: beastie.saturn-tech.com: drussell owned process doing -bs Date: Thu, 21 Jun 2001 15:46:12 -0600 (MDT) From: Doug Russell To: freebsd-smp@FreeBSD.ORG Subject: Re: clock speedup on SMP boxes In-Reply-To: <7314.993066048@critter> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I seem to be seeing the same problem here on one of my machines, but it is running UP. There are 4 PPROs installed in the box, but I've never gotten around to figuring out why it won't boot SMP. See my next message with mptable output, etc. This machine gains about 2000 seconds per day: FreeBSD beastie.saturn-tech.com 4.2-STABLE FreeBSD 4.2-STABLE #1: Tue Nov 21 13:59:23 MST 2000 \ drussell@beastie.saturn-tech.com:/usr/src/sys/compile/BEASTIE i386. 3:38PM up 82 days, 20:30, 3 users, load averages: 0.01, 0.00, 0.00 While, this one, running UP (2 processors installed) running a NEWER world, is not currently exhibiting this behavior: FreeBSD mickey.saturn-tech.com 4.2-STABLE FreeBSD 4.2-STABLE #0: \ Thu Jan 11 17:17:23 MST 2001 \ drussell@mickey.saturn-tech.com:/usr/src/sys/compile/MICKEY i386 3:36PM up 55 days, 9:17, 1 user, load averages: 0.00, 0.22, 0.27 I haven't had a chance to move things around and reboot or update the world on these machines, but I have at least two more working NF9000s I can use to test with, if it helps, but they take about 10 minutes to boot with 512M-1G of RAM, which makes it a bit of a pain. :) I was hoping to have a few ideas before I spend hours and hours twiddling. > I don't have a good idea right now what causes this, but I have a number > of ideas I would like people to try out: ... I'll give these suggestions from phk a whirl when I get a chance, assuming this really is the same problem I'm seeing here. Later...... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jun 21 15: 2:28 2001 Delivered-To: freebsd-smp@freebsd.org Received: from beastie.saturn-tech.com (beastie.saturn-tech.com [207.229.19.136]) by hub.freebsd.org (Postfix) with ESMTP id 5FFC337B401 for ; Thu, 21 Jun 2001 15:02:17 -0700 (PDT) (envelope-from drussell@saturn-tech.com) Received: from localhost (drussell@localhost) by beastie.saturn-tech.com (8.11.1/8.11.0) with ESMTP id f5LM73r43695 for ; Thu, 21 Jun 2001 16:07:03 -0600 (MDT) (envelope-from drussell@saturn-tech.com) X-Authentication-Warning: beastie.saturn-tech.com: drussell owned process doing -bs Date: Thu, 21 Jun 2001 16:07:03 -0600 (MDT) From: Doug Russell To: freebsd-smp@FreeBSD.org Subject: Netframe NF9000 series won't boot SMP Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I have several old NetFrame NF9016 systems (actually a couple are actually NEW from when NetFrame closed their spares warehouse) and spare MBs, PCI cans, power supplies, etc. These machines have a bit of a strange PCI-PCI-PCI bus architecture with each canister of 4 PCI slots being behind a PCI-PCI bridge, etc. For this reason, I'm not surprised they don't want to boot properly out-of-the-box. The only real BIOS options for SMP are MP table version 1.1 / 1.4, and an option for Sequential/Sparse PCI bus allocation, which just seems to change the way the busses are numbered in the table, to no avail. Sequential seems to make the most sense numerically, and causes FreeBSD to find devices plugged into all canisters properly. NT4 requires the Sparse option, and works with 4 processors and all PCI busses, but W2K is known to not support this PCI configuration at all. It finds the local Adaptec SCSI, the 5436 video controller, and that's it. I beleive this is an IRQ routing problem, as the machine crashes at the Routing Intpin ... section of the boot. Forgive the poor information, this is the first SMP machine I've ever had that didn't Just Work. :) Hopefully someone can spot the bogusness. :) mptable -verbose with BIOS set to Sequential, version 1.1 of the MP spec: =============================================================================== MPTable, version 2.0.15 looking for EBDA pointer @ 0x040e, found, searching EBDA @ 0x0009fc00 searching CMOS 'top of mem' @ 0x0009f800 (638K) searching default 'top of mem' @ 0x0009fc00 (639K) searching BIOS @ 0x000f0000 MP FPS found in BIOS @ physical addr: 0x000fbd10 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000fbd10 signature: '_MP_' length: 16 bytes version: 1.1 checksum: 0x4d mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x0009fd50 signature: 'PCMP' base table length: 468 version: 1.1 checksum: 0x47 OEM ID: 'INTEL ' Product ID: 'ORION ' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 47 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 3 0x11 BSP, usable 6 1 9 0xfbff 0 0x11 AP, usable 6 1 9 0xfbff 1 0x11 AP, usable 6 1 9 0xfbff 2 0x11 AP, usable 6 1 9 0xfbff -- Bus: Bus ID Type 0 PCI 1 PCI 2 PCI 3 PCI 4 PCI 5 PCI 6 PCI 176 ISA -- I/O APICs: APIC ID Version State Address 4 0x11 usable 0xfec00000 5 0x11 usable 0xfec01000 6 0x11 usable 0xfec02000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 176 2 4 2 INT active-hi edge 176 1 4 1 INT active-hi edge 176 0 4 0 INT active-hi edge 176 3 4 3 INT active-hi edge 176 4 4 4 INT active-hi edge 176 5 4 5 INT active-hi edge 176 6 4 6 INT active-hi edge 176 7 4 7 INT active-hi edge 176 8 4 8 INT active-hi edge 176 9 4 9 INT active-hi edge 176 12 4 12 INT active-hi edge 176 13 4 13 INT active-hi edge 176 15 4 15 INT active-hi level 0 3:A 4 14 INT active-hi level 2 4:A 5 3 INT active-hi level 2 5:A 5 4 INT active-hi level 2 6:A 5 5 INT active-hi level 2 7:A 5 6 INT active-hi level 3 4:A 5 9 INT active-hi level 3 5:A 5 10 INT active-hi level 3 6:A 5 11 INT active-hi level 3 7:A 5 12 INT active-hi level 5 4:A 6 3 INT active-hi level 5 5:A 6 4 INT active-hi level 5 6:A 6 5 INT active-hi level 5 7:A 6 6 INT active-hi level 6 4:A 6 9 INT active-hi level 6 5:A 6 10 INT active-hi level 6 6:A 6 11 INT active-hi level 6 7:A 6 12 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 176 0 255 0 NMI active-hi edge 176 0 255 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=4 # number of CPUs #options NBUS=8 # number of busses #options NAPIC=3 # number of IO APICs #options NINTR=30 # number of INTs =============================================================================== MP table with BIOS set to v1.4, again sequential: =============================================================================== MPTable, version 2.0.15 looking for EBDA pointer @ 0x040e, found, searching EBDA @ 0x0009fc00 searching CMOS 'top of mem' @ 0x0009f800 (638K) searching default 'top of mem' @ 0x0009fc00 (639K) searching BIOS @ 0x000f0000 MP FPS found in BIOS @ physical addr: 0x000fbd10 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000fbd10 signature: '_MP_' length: 16 bytes version: 1.4 checksum: 0x4a mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x0009fd50 signature: 'PCMP' base table length: 468 version: 1.4 checksum: 0x65 OEM ID: 'INTEL ' Product ID: 'ORION ' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 47 local APIC address: 0xfee00000 extended table length: 164 extended table checksum: 59 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 3 0x11 BSP, usable 6 1 9 0xfbff 0 0x11 AP, usable 6 1 9 0xfbff 1 0x11 AP, usable 6 1 9 0xfbff 2 0x11 AP, usable 6 1 9 0xfbff -- Bus: Bus ID Type 0 PCI 1 PCI 2 PCI 3 PCI 4 PCI 5 PCI 6 PCI 176 ISA -- I/O APICs: APIC ID Version State Address 4 0x11 usable 0xfec00000 5 0x11 usable 0xfec01000 6 0x11 usable 0xfec02000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 176 2 4 2 INT active-hi edge 176 1 4 1 INT active-hi edge 176 0 4 0 INT active-hi edge 176 3 4 3 INT active-hi edge 176 4 4 4 INT active-hi edge 176 5 4 5 INT active-hi edge 176 6 4 6 INT active-hi edge 176 7 4 7 INT active-hi edge 176 8 4 8 INT active-hi edge 176 9 4 9 INT active-hi edge 176 12 4 12 INT active-hi edge 176 13 4 13 INT active-hi edge 176 15 4 15 INT active-hi level 0 3:A 4 14 INT active-hi level 2 4:A 5 3 INT active-hi level 2 5:A 5 4 INT active-hi level 2 6:A 5 5 INT active-hi level 2 7:A 5 6 INT active-hi level 3 4:A 5 9 INT active-hi level 3 5:A 5 10 INT active-hi level 3 6:A 5 11 INT active-hi level 3 7:A 5 12 INT active-hi level 5 4:A 6 3 INT active-hi level 5 5:A 6 4 INT active-hi level 5 6:A 6 5 INT active-hi level 5 7:A 6 6 INT active-hi level 6 4:A 6 9 INT active-hi level 6 5:A 6 10 INT active-hi level 6 6:A 6 11 INT active-hi level 6 7:A 6 12 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 176 0 255 0 NMI active-hi edge 176 0 255 1 ------------------------------------------------------------------------------- MP Config Extended Table Entries: -- System Address Space bus ID: 0 address type: memory address address base: 0x0 address range: 0x80000000 -- System Address Space bus ID: 0 address type: memory address address base: 0xe0000000 address range: 0xfcbfffff -- System Address Space bus ID: 0 address type: I/O address address base: 0x0 address range: 0xffff -- System Address Space bus ID: 1 address type: memory address address base: 0xfec00000 address range: 0x1000 -- System Address Space bus ID: 1 address type: memory address address base: 0xa0000 address range: 0x20000 -- System Address Space bus ID: 1 address type: memory address address base: 0xf0000 address range: 0x10000 -- System Address Space bus ID: 1 address type: memory address address base: 0xffe00000 address range: 0x200000 -- Bus Heirarchy bus ID: 176 bus info: 0x01 parent bus ID: 0 -- Compatibility Bus Address bus ID: 176 address modifier: add predefined range: 0x00000000 -- Compatibility Bus Address bus ID: 176 address modifier: add predefined range: 0x00000001 ------------------------------------------------------------------------------- # 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=4 # number of CPUs #options NBUS=8 # number of busses #options NAPIC=3 # number of IO APICs #options NINTR=30 # number of INTs =============================================================================== Sparse mode simply changes the bus numbering of the PCI busses from: 3 ,4 ,5 ,6 to 35, 128, 130, 163 Any ideas would be greatly appreciated! Later...... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message