From owner-freebsd-smp Mon Mar 3 9:43:53 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4312137B401 for ; Mon, 3 Mar 2003 09:43:50 -0800 (PST) Received: from sinamail.com (61-221-29-145.HINET-IP.hinet.net [61.221.29.145]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9A4F543FBF for ; Mon, 3 Mar 2003 09:43:48 -0800 (PST) (envelope-from suppergeorge@sinamail.com) From: star@yahoo.com.tw To: smp@FreeBSD.org Subject: =?ISO-8859-1?B?prO+97d8p0HEQLdOpWi5wbjVttw/Pw==?= Reply-To: suppergeorge@sinamail.com Date: 04 Mar 2003 01:48:56 +0800 MIME-Version: 1.0 Content-Type: text/html Content-Transfer-Encoding: 8bit Message-Id: <20030303174348.9A4F543FBF@mx1.FreeBSD.org> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org 為什麼有人會比你成功10倍
主旨: 這或許是您正在尋找的機會喔
這或許是您正在找的機會哦！
對不起！打擾了，如果因此造成您的困擾，請直接刪除本信及點選下方「不想再收信」，我們會將您的資料刪除！
為什麼有人會比你成功10倍，收入多100倍、甚至多1000倍，難道他有比你多聰明這麼多嗎？
答案肯定不是的！
想一想！那些收入比我們高很多，生活比我們好很多的人！
他們到底做了什麼是我們所不知道的事？
而我們到底做錯了什麼、又錯過了什麼？
想不想知道人家怎麼做倒的！
你相信「時間＝金錢」、還是「時間＞金錢」

舉例：

我們一天工作8小時，一年工作365天，一輩子工作30年！那我們一輩子的總工作時數？
８小時＊３６５天＊３０年＝８７,６００小時
如果你的時薪100元，你一輩子賺876萬元！
如果你的時薪150元，你一輩子賺1314萬元！
如果你的時薪200元，你一輩子賺1752萬元！
看起來好像很多，看清楚！一年工作365天，要工作30年！而且不吃不喝！
這樣的收入，足夠三餐溫飽；買車子、房子勉強夠用；別忘了，還有子女的教育費、自己的養老金、還有『夢想』等待實現！
這樣的一輩子，你甘心嗎？
身為員工的你，每天辛苦為的是什麼？家庭、小孩？你有沒有想過，你上班一輩子，將來你的小孩能承接你的職位繼續做下去嗎？（除非你自己是老闆）
想不想改變自己及下一代的一生？
不想再收信(Unsubscribe)

　

To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 3 12:28:48 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8720937B401 for ; Mon, 3 Mar 2003 12:28:47 -0800 (PST) Received: from sccrmhc01.attbi.com (sccrmhc01.attbi.com [204.127.202.61]) by mx1.FreeBSD.org (Postfix) with ESMTP id 784A843FB1 for ; Mon, 3 Mar 2003 12:28:46 -0800 (PST) (envelope-from julian@elischer.org) Received: from interjet.elischer.org (12-232-168-4.client.attbi.com[12.232.168.4]) by sccrmhc01.attbi.com (sccrmhc01) with ESMTP id <2003030320284500100j6e8oe>; Mon, 3 Mar 2003 20:28:45 +0000 Received: from localhost (localhost.elischer.org [127.0.0.1]) by InterJet.elischer.org (8.9.1a/8.9.1) with ESMTP id MAA43295 for ; Mon, 3 Mar 2003 12:28:44 -0800 (PST) Date: Mon, 3 Mar 2003 12:28:43 -0800 (PST) From: Julian Elischer To: smp@freebsd.org Subject: UP ans SMP on the same kernel 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 How much work would it take to make a kernel that could run on both SMP and UP machines (and use the extra CPUs?) UP kernels can run on SMP hardware, so I guess that it's mainly finding a way of switching from using the PIC to using some APICS dynamically, when you find that you have SMP hardware. Incidently I have several UP machines that have APICs and it may be effective to be able to run an APIC kernel on this hardware.. (This all assumes that the rest of the kernel was compiled SMP to have all the lockeng etc.) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 3 12:55:12 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 771FC37B401 for ; Mon, 3 Mar 2003 12:55:11 -0800 (PST) Received: from smtp1.Stanford.EDU (smtp1.Stanford.EDU [171.64.14.23]) by mx1.FreeBSD.org (Postfix) with ESMTP id 085CC43FE1 for ; Mon, 3 Mar 2003 12:55:11 -0800 (PST) (envelope-from jstockdale@stanford.edu) Received: from smtp1.Stanford.EDU ([127.0.0.1]) by smtp1.Stanford.EDU (8.12.6/8.12.6) with ESMTP id h23Kt9HI005633 for ; Mon, 3 Mar 2003 12:55:09 -0800 (PST) Received: from stanford.edu (mentat.Stanford.EDU [128.12.44.75]) by smtp1.Stanford.EDU (8.12.6/8.12.6) with ESMTP id h23Kt8DS005625 for ; Mon, 3 Mar 2003 12:55:09 -0800 (PST) Date: Mon, 3 Mar 2003 12:55:15 -0800 Mime-Version: 1.0 (Apple Message framework v551) Content-Type: text/plain; charset=US-ASCII; format=flowed Subject: Tyan i7505 + Dual Xeon: Issues with hang during APIC IO initialization during boot. From: John Stockdale To: smp@freebsd.org Content-Transfer-Encoding: 7bit Message-Id: <6F074D8E-4DBA-11D7-8D9A-000393A6EB58@stanford.edu> X-Mailer: Apple Mail (2.551) 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'm building a FreeBSD system based on the Tyan Thunder i7505 motherboard, with Dual 2GHz, 533 Mhz FSB Xeon processors. Although the UP kernel boots just fine, when I recompile for SMP and APIC support the system locks up right after it displays: "APIC_IO: Testing 8254 interrupt delivery" I've seen some comments previously on the list but most of them seem to either have been resolved in either the newer implementations of the kernel or by the motherboard manufactures themselves. I've already emailed the Tyan support people with this query, but I was wondering if anyone else had any suggestions. The motherboard does have bios settings for OS/HT/MP modes and I have tried the following combinations: WinXP/2000; HT Enabled; MP 1.4 Other/Linux; HT Enabled; MP 1.4 Other/Linux; HT Disabled; MP 1.4 Other Linux; HT Enabled; MP 1.1 And have not had any change in behavior under any of these settings. On a related note the UP kernel boots great and everything runs well, but obviously I would like to get the SMP support working so that one proc isn't just idling. Additionally, can anyone tell me how to view the system information status (ie. proc status/memory/hdd/etc) in FreeBSD. I'm sure I should know this but this is my first real FBSD machine, and I havn't been using Unix for awhile. Thanks John Stockdale jstockdale@stanford.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 3 15: 5:42 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 091AA37B509 for ; Mon, 3 Mar 2003 15:05:41 -0800 (PST) Received: from mail.speakeasy.net (mail13.speakeasy.net [216.254.0.213]) by mx1.FreeBSD.org (Postfix) with ESMTP id BC05E43F75 for ; Mon, 3 Mar 2003 15:05:39 -0800 (PST) (envelope-from jhb@FreeBSD.org) Received: (qmail 5682 invoked from network); 3 Mar 2003 23:05:47 -0000 Received: from unknown (HELO server.baldwin.cx) ([216.27.160.63]) (envelope-sender ) by mail13.speakeasy.net (qmail-ldap-1.03) with DES-CBC3-SHA encrypted SMTP for ; 3 Mar 2003 23:05:47 -0000 Received: from laptop.baldwin.cx (gw1.twc.weather.com [216.133.140.1]) by server.baldwin.cx (8.12.6/8.12.6) with ESMTP id h23N3IhT042723; Mon, 3 Mar 2003 18:03:18 -0500 (EST) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.5.2 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: Mon, 03 Mar 2003 18:05:53 -0500 (EST) From: John Baldwin To: Julian Elischer Subject: RE: UP ans SMP on the same kernel 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 03-Mar-2003 Julian Elischer wrote: > > > How much work would it take to make a kernel that could run > on both SMP and UP machines (and use the extra CPUs?) > > UP kernels can run on SMP hardware, so I guess that it's mainly finding > a way of switching from using the PIC to using some APICS > dynamically, when you find that you have SMP hardware. > > Incidently I have several UP machines that have APICs > and it may be effective to be able to run an APIC > kernel on this hardware.. > > (This all assumes that the rest of the kernel was compiled > SMP to have all the lockeng etc.) I am working on this, but it's taking a while amongst other distractions. The real work is in the PIC vs. APIC stuff. The actual SMP part is fairly easy as the sparc64, alpha, and ia64 ports demonstrate. -- John Baldwin <>< http://www.FreeBSD.org/~jhb/ "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 Mon Mar 3 15:52: 1 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id EB3BB37B401 for ; Mon, 3 Mar 2003 15:51:59 -0800 (PST) Received: from archway.securecomputing.com (archway.securecomputing.com [192.55.214.7]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1570743F3F for ; Mon, 3 Mar 2003 15:51:59 -0800 (PST) (envelope-from minear@securecomputing.com) Date: Mon, 3 Mar 2003 17:51:52 -0600 (CST) Message-Id: <200303032351.h23Npq029060@sem.sctc.com> From: Spencer Minear To: freebsd-smp@freebsd.org Subject: Compaq 370/380/580 G3 SMP problem details? 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'm told that there may have been some discussion in a FreeBSD mail list about the real details behind the problems with new Compaq G3 servers not running in MP mode. The root problem appears to be that on these systems, placing a value of 0x10 in the Local APIC TPR register results in blocking all interrupts. The spec clearly says that this should not block interrupts with priorty greater than 1, but it sure appears to be happening. As mentioned I've been lead to belive that there are some entries in the FreeBSD mail archive discussing the details behind the problem, but I cannot find them. I have had one customer report that upgrading the BIOS on a 370 to the lastest reve resolved the problem on that system. I have upgraded the BIOS on the 380 to the lastest available from Compaq and found that did not provide any change in behavior. If anyone is can point me to the details that may have been discussed or have more details on the solutions available from Compaq I would appreciate hearing from you. Thanks, -- Spence Minear To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 3 16: 8:37 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8AC1E37B401; Mon, 3 Mar 2003 16:08:36 -0800 (PST) Received: from sccrmhc03.attbi.com (sccrmhc03.attbi.com [204.127.202.63]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8ECCA43F93; Mon, 3 Mar 2003 16:08:35 -0800 (PST) (envelope-from julian@elischer.org) Received: from interjet.elischer.org (12-232-168-4.client.attbi.com[12.232.168.4]) by sccrmhc03.attbi.com (sccrmhc03) with ESMTP id <2003030400083400300aj3tje>; Tue, 4 Mar 2003 00:08:34 +0000 Received: from localhost (localhost.elischer.org [127.0.0.1]) by InterJet.elischer.org (8.9.1a/8.9.1) with ESMTP id QAA44878; Mon, 3 Mar 2003 16:08:32 -0800 (PST) Date: Mon, 3 Mar 2003 16:08:31 -0800 (PST) From: Julian Elischer To: John Baldwin Cc: smp@freebsd.org Subject: RE: UP ans SMP on the same kernel In-Reply-To: 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 Mon, 3 Mar 2003, John Baldwin wrote: > > On 03-Mar-2003 Julian Elischer wrote: > > > > > > How much work would it take to make a kernel that could run > > on both SMP and UP machines (and use the extra CPUs?) > > > > UP kernels can run on SMP hardware, so I guess that it's mainly finding > > a way of switching from using the PIC to using some APICS > > dynamically, when you find that you have SMP hardware. > > > > Incidently I have several UP machines that have APICs > > and it may be effective to be able to run an APIC > > kernel on this hardware.. > > > > (This all assumes that the rest of the kernel was compiled > > SMP to have all the lockeng etc.) > > I am working on this, but it's taking a while amongst other > distractions. The real work is in the PIC vs. APIC stuff. > The actual SMP part is fairly easy as the sparc64, alpha, and > ia64 ports demonstrate. That's good to know.. We have spent a lot of effort making one "universal" kernel for our systems in the field, but now that we have some SMP machines, suddenly we have 2 again. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 3 16:29: 1 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0183837B401 for ; Mon, 3 Mar 2003 16:29:00 -0800 (PST) Received: from ztxmail03.ztx.compaq.com (ztxmail03.ztx.compaq.com [161.114.1.207]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1D27843F3F for ; Mon, 3 Mar 2003 16:28:59 -0800 (PST) (envelope-from john.cagle@hp.com) Received: from cceexg13.americas.cpqcorp.net (cceexg13.americas.cpqcorp.net [16.110.250.119]) by ztxmail03.ztx.compaq.com (Postfix) with ESMTP id 118E252EF; Mon, 3 Mar 2003 18:28:58 -0600 (CST) Received: from cceexc19.americas.cpqcorp.net ([16.110.250.85]) by cceexg13.americas.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.2966); Mon, 3 Mar 2003 18:28:57 -0600 X-MimeOLE: Produced By Microsoft Exchange V6.0.6375.0 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Subject: RE: Compaq 370/380/580 G3 SMP problem details? Date: Mon, 3 Mar 2003 18:28:57 -0600 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: Compaq 370/380/580 G3 SMP problem details? Thread-Index: AcLh3+Zvqbqm3YokST2ImNkPPW4sgQAAzkkw From: "Cagle, John (ISS-Houston)" To: "Spencer Minear" Cc: X-OriginalArrivalTime: 04 Mar 2003 00:28:57.0923 (UTC) FILETIME=[0BA9C130:01C2E1E5] 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 Spencer, Peter Wemm checked in a new version of mpapic.c around the end of January to fix this issue. I have verified that SMP with RELENG_4 now works with our latest G3 architecture servers, including hyperthreading. Which version of FreeBSD are you using? Also, can you point me to the spec that discusses the TPR? I haven't been able to find that one... Thanks, John -------------------------------- John Cagle john.cagle@hp.com Principal Member Technical Staff Industry Standard Servers Hewlett-Packard Company > -----Original Message----- > From: Spencer Minear [mailto:spencer_minear@securecomputing.com]=20 > Sent: Monday, March 03, 2003 5:52 PM > To: freebsd-smp@freebsd.org > Subject: Compaq 370/380/580 G3 SMP problem details? >=20 >=20 >=20 > I'm told that there may have been some discussion in a=20 > FreeBSD mail list about the real details behind the problems=20 > with new Compaq G3 servers not running in MP mode. The root=20 > problem appears to be that on these systems, placing a value=20 > of 0x10 in the Local APIC TPR register results in blocking=20 > all interrupts. The spec clearly says that this should not=20 > block interrupts with priorty greater than 1, but it sure=20 > appears to be happening. >=20 > As mentioned I've been lead to belive that there are some=20 > entries in the FreeBSD mail archive discussing the details=20 > behind the problem, but I cannot find them. I have had one=20 > customer report that upgrading the BIOS on a 370 to the=20 > lastest reve resolved the problem on that system. I have=20 > upgraded the BIOS on the 380 to the lastest available from=20 > Compaq and found that did not provide any change in behavior. >=20 > If anyone is can point me to the details that may have been=20 > discussed or have more details on the solutions available=20 > from Compaq I would appreciate hearing from you. >=20 > Thanks, >=20 > -- Spence Minear To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 3 16:31:49 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0657537B401 for ; Mon, 3 Mar 2003 16:31:48 -0800 (PST) Received: from pirx.hexapodia.org (pirx.hexapodia.org [208.42.114.113]) by mx1.FreeBSD.org (Postfix) with ESMTP id D814943F3F for ; Mon, 3 Mar 2003 16:31:46 -0800 (PST) (envelope-from adi@hexapodia.org) Received: by pirx.hexapodia.org (Postfix, from userid 22448) id 3DF4CB404; Mon, 3 Mar 2003 18:31:45 -0600 (CST) Date: Mon, 3 Mar 2003 18:31:45 -0600 From: Andy Isaacson To: Spencer Minear Cc: freebsd-smp@freebsd.org Subject: Re: Compaq 370/380/580 G3 SMP problem details? Message-ID: <20030303183144.A18225@hexapodia.org> References: <200303032351.h23Npq029060@sem.sctc.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <200303032351.h23Npq029060@sem.sctc.com>; from spencer_minear@securecomputing.com on Mon, Mar 03, 2003 at 05:51:52PM -0600 X-PGP-Fingerprint: 48 01 21 E2 D4 E4 68 D1 B8 DF 39 B2 AF A3 16 B9 X-PGP-Key-URL: http://web.hexapodia.org/~adi/pgp.txt X-Domestic-Surveillance: money launder bomb tax evasion 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 Mon, Mar 03, 2003 at 05:51:52PM -0600, Spencer Minear wrote: > I'm told that there may have been some discussion in a FreeBSD mail > list about the real details behind the problems with new Compaq G3 > servers not running in MP mode. The root problem appears to be that > on these systems, placing a value of 0x10 in the Local APIC TPR > register results in blocking all interrupts. The spec clearly says > that this should not block interrupts with priorty greater than 1, but > it sure appears to be happening. > > As mentioned I've been lead to belive that there are some entries in > the FreeBSD mail archive discussing the details behind the problem, > but I cannot find them. I have had one customer report that upgrading > the BIOS on a 370 to the lastest reve resolved the problem on that > system. I have upgraded the BIOS on the 380 to the lastest available > from Compaq and found that did not provide any change in behavior. > > If anyone is can point me to the details that may have been discussed > or have more details on the solutions available from Compaq I would > appreciate hearing from you. See the thread "4.7 on 2 Xeons SMP?" dated January 22. Message-ID: > Paul Saab writes: > > >> By the way, this is also a problem with 5.0-release. > > > > Peter committed a fix to -current solves this problem. a RELENG_4 > > commit > > is coming shortly to fix it there as well. > > Yep ! I updated (-CURRENT) my Proliant ML370 this morning, and the fix > works well. With hypethreading enabled, FreeBSD now uses 4 CPUs. :-) > > You all did a really great job ! > Thanks a lot. -andy To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 3 21:19:34 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C713937B401 for ; Mon, 3 Mar 2003 21:19:32 -0800 (PST) Received: from smtp018.mail.yahoo.com (smtp018.mail.yahoo.com [216.136.174.115]) by mx1.FreeBSD.org (Postfix) with SMTP id BE50943FB1 for ; Mon, 3 Mar 2003 21:19:31 -0800 (PST) (envelope-from kaeru@pd.jaring.my) Received: from unknown (HELO ?219.93.86.234?) (khairil?yusof@219.93.86.234 with plain) by smtp.mail.vip.sc5.yahoo.com with SMTP; 4 Mar 2003 05:19:31 -0000 Subject: Temporary lockups on 5-release From: Khairil Yusof Reply-To: kaeru@pd.jaring.my To: freebsd-smp@FreeBSD.ORG Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-c5DWpRHG8h/OfCWkpFaq" Organization: Message-Id: <1046755156.598.92.camel@daemon.home.net> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.2 Date: 04 Mar 2003 13:19:18 +0800 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 --=-c5DWpRHG8h/OfCWkpFaq Content-Type: text/plain Content-Transfer-Encoding: quoted-printable Since upgrading to 5.0, I often get temporary lockups. Basically the system just hangs for a second or two, almost as if it's going to freeze and hang, then continues again. I'm no expert on SMP, but how can a user like me, help out trace problems like this? Or is this behaviour expected due to SMPng not being complete and a lot of drivers still using Giant lock? --=20 Khairil Yusof --=-c5DWpRHG8h/OfCWkpFaq Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (FreeBSD) iD8DBQA+ZDdTDAqnLW/+/X8RAiudAJ4nDpM2axg75IYtrnjIzrfS3pA6RACgkkmL 10RmmHPqD54xVserzyXikd0= =cBow -----END PGP SIGNATURE----- --=-c5DWpRHG8h/OfCWkpFaq-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 3 23:40:39 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id AB00337B401 for ; Mon, 3 Mar 2003 23:40:37 -0800 (PST) Received: from mx0.gmx.net (mx0.gmx.net [213.165.64.100]) by mx1.FreeBSD.org (Postfix) with SMTP id EC21443FA3 for ; Mon, 3 Mar 2003 23:40:33 -0800 (PST) (envelope-from pcc@gmx.net) Received: (qmail 5835 invoked by uid 0); 4 Mar 2003 07:40:32 -0000 Date: Tue, 4 Mar 2003 08:40:32 +0100 (MET) From: Peter Cornelius To: Julian Elischer Cc: smp@freebsd.org MIME-Version: 1.0 References: <3821.1046762486@www54.gmx.net> Subject: Re: UP ans SMP on the same kernel X-Priority: 4 (Low) X-Authenticated-Sender: #0000491680@gmx.net X-Authenticated-IP: [212.86.205.70] Message-ID: <16352.1046763632@www54.gmx.net> X-Mailer: WWW-Mail 1.6 (Global Message Exchange) X-Flags: 0001 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit 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 Reply-To: pcc@gmx.net,smp@freebsd.org > How much work would it take to make a kernel that could run > on both SMP and UP machines (and use the extra CPUs?) I have been able to run -STABLE (last cvsup somewhere a week or two ago) with SMP enabled for more than half a year now, I think, on a UP machine since I noted that Sandra on Windows told me it was an MP 1.4-compliant MB (did I ever run windows since, I can't remember...). So I don't know if that is a significant waypoint. I run the box diskless making installworlds and such a bit tricky at times which is why I don't upgrade too often, though (and we also just moved making things even trickier ;-)). It really is a cheapo all-in-there-complete box from the non-food heap of a food chain supermarket which I did not even get a proper main board manual for but I have been made to believe it's a MicroStar? MS-6399 (???) or something like that (I'd have to check when I'm back home if that is of interest). However, the board manufacturer only point to the distributor and they only point to their press-this-key-here-to-start-windows 'handbook'. (Sigh) > UP kernels can run on SMP hardware, so I guess that it's mainly finding > a way of switching from using the PIC to using some APICS > dynamically, when you find that you have SMP hardware. > > Incidently I have several UP machines that have APICs > and it may be effective to be able to run an APIC > kernel on this hardware.. > > (This all assumes that the rest of the kernel was compiled > SMP to have all the lockeng etc.) Anyways, hope this helps. Best regards, Peter. -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte l踄heln! Fotogalerie online mit GMX ohne eigene Homepage! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 3 23:48: 3 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 589F137B401 for ; Mon, 3 Mar 2003 23:48:02 -0800 (PST) Received: from obsecurity.dyndns.org (adsl-63-207-60-52.dsl.lsan03.pacbell.net [63.207.60.52]) by mx1.FreeBSD.org (Postfix) with ESMTP id 983CE43FAF for ; Mon, 3 Mar 2003 23:48:01 -0800 (PST) (envelope-from kris@obsecurity.org) Received: from rot13.obsecurity.org (rot13.obsecurity.org [10.0.0.5]) by obsecurity.dyndns.org (Postfix) with ESMTP id 4369066B37; Mon, 3 Mar 2003 23:48:01 -0800 (PST) Received: by rot13.obsecurity.org (Postfix, from userid 1000) id 22B001006; Mon, 3 Mar 2003 23:48:01 -0800 (PST) Date: Mon, 3 Mar 2003 23:48:00 -0800 From: Kris Kennaway To: Khairil Yusof Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Temporary lockups on 5-release Message-ID: <20030304074800.GA89765@rot13.obsecurity.org> References: <1046755156.598.92.camel@daemon.home.net> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="3MwIy2ne0vdjdPXF" Content-Disposition: inline In-Reply-To: <1046755156.598.92.camel@daemon.home.net> User-Agent: Mutt/1.4i 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 --3MwIy2ne0vdjdPXF Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Mar 04, 2003 at 01:19:18PM +0800, Khairil Yusof wrote: > Since upgrading to 5.0, I often get temporary lockups. >=20 > Basically the system just hangs for a second or two, almost as if it's > going to freeze and hang, then continues again. >=20 > I'm no expert on SMP, but how can a user like me, help out trace > problems like this? >=20 > Or is this behaviour expected due to SMPng not being complete and a lot > of drivers still using Giant lock? It can be due to Giant contention..was the system under heavy load at the time? Kris --3MwIy2ne0vdjdPXF Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (FreeBSD) iD8DBQE+ZFowWry0BWjoQKURAraPAJ0Q2XY8E1uxpLrNgw9XGz+Gmi96sACgoV/y jiKWl5wp6wh1G1QkYeTu2b0= =8F/p -----END PGP SIGNATURE----- --3MwIy2ne0vdjdPXF-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Mar 4 6: 8: 5 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4065137B401 for ; Tue, 4 Mar 2003 06:08:03 -0800 (PST) Received: from smtp012.mail.yahoo.com (smtp012.mail.yahoo.com [216.136.173.32]) by mx1.FreeBSD.org (Postfix) with SMTP id 4B4A343F85 for ; Tue, 4 Mar 2003 06:08:02 -0800 (PST) (envelope-from kaeru@pd.jaring.my) Received: from unknown (HELO ?219.93.86.251?) (khairil?yusof@219.93.86.251 with plain) by smtp.mail.vip.sc5.yahoo.com with SMTP; 4 Mar 2003 14:08:01 -0000 Subject: Re: Temporary lockups on 5-release From: Khairil Yusof Reply-To: kaeru@pd.jaring.my To: Kris Kennaway Cc: freebsd-smp@FreeBSD.ORG In-Reply-To: <20030304074800.GA89765@rot13.obsecurity.org> References: <1046755156.598.92.camel@daemon.home.net> <20030304074800.GA89765@rot13.obsecurity.org> Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-LRDZfzOG0DuLS4Y4zQ73" Organization: Message-Id: <1046786866.578.197.camel@daemon.home.net> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.2 Date: 04 Mar 2003 22:07:48 +0800 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 --=-LRDZfzOG0DuLS4Y4zQ73 Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Tue, 2003-03-04 at 15:48, Kris Kennaway wrote: > It can be due to Giant contention..was the system under heavy load at > the time? =46rom what I'm observing, the jarring lockups (more than a second) can also occur during a one time memory and cpu intensive operation.=20 Examples being, cut and pasting portions of a 20MB image with gimp, changing desktop background (1280x1024x32) etc. These are not swap related (384MB and almost 0 swap usage). It can also be repeated when using large apps like mozilla or nautilus. This is one example, though it can also happen during intensive disk operations (this may be due more to IDE), so I'll look into this aspect later to see if there is a correlation when these lockups appear with disk operations while removing the IDE bottleneck aspect. These lockups are worrying for me because, they lead to system reboots. And they are my personal benchmark for stability (ie can FreeBSD recover or not when it happens). FreeBSD 5 RC-1 -> system reboots FreeBSD 5 RC-2 -> ok FREEBSD 5 RC-3 and RELEASE -> ok FREEBSD CURRENT (as of last weekend) -> system reboot. I'm back to RELEASE, because the constant reboots can cause data corruption which is not good for my one and only home system. --=20 Khairil Yusof --=-LRDZfzOG0DuLS4Y4zQ73 Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (FreeBSD) iD8DBQA+ZLMyDAqnLW/+/X8RAse4AKC2KQ07S+HIoyB7kFCykHhdDJ3sWQCg3d3P a+S7oSZ0Rna9H+gFyyfwrak= =Woll -----END PGP SIGNATURE----- --=-LRDZfzOG0DuLS4Y4zQ73-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Mar 4 6:35: 1 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 35C2437B401 for ; Tue, 4 Mar 2003 06:35:00 -0800 (PST) Received: from archway.securecomputing.com (archway.securecomputing.com [192.55.214.7]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5BF7543FBD for ; Tue, 4 Mar 2003 06:34:59 -0800 (PST) (envelope-from minear@securecomputing.com) Date: Tue, 4 Mar 2003 08:34:51 -0600 (CST) Message-Id: <200303041434.h24EYp829138@sem.sctc.com> From: Spencer Minear To: john.cagle@hp.com Cc: spencer_minear@securecomputing.com, freebsd-smp@freebsd.org In-reply-to: (john.cagle@hp.com) Subject: RE: Compaq 370/380/580 G3 SMP problem details? 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 found it in chapter 8 of the Ia-32 Intel Architecture Software Developers manual. It can be found the the Intel developer web site I think http://www.intel.com/design/pentium4/manuals/245471.htm provides links to buy a hardcopy or down load a soft copy. > Which version of FreeBSD are you using? Actually I'm cheeting, we use BSD/OS along with some FreeBSD in our developement group. -- Spence To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Mar 4 12:33:57 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8652637B401 for ; Tue, 4 Mar 2003 12:33:55 -0800 (PST) Received: from obsecurity.dyndns.org (adsl-63-207-60-52.dsl.lsan03.pacbell.net [63.207.60.52]) by mx1.FreeBSD.org (Postfix) with ESMTP id C3D8443F3F for ; Tue, 4 Mar 2003 12:33:54 -0800 (PST) (envelope-from kris@obsecurity.org) Received: from rot13.obsecurity.org (rot13.obsecurity.org [10.0.0.5]) by obsecurity.dyndns.org (Postfix) with ESMTP id 7880E66B60; Tue, 4 Mar 2003 12:33:53 -0800 (PST) Received: by rot13.obsecurity.org (Postfix, from userid 1000) id 5ED8D1023; Tue, 4 Mar 2003 12:33:53 -0800 (PST) Date: Tue, 4 Mar 2003 12:33:53 -0800 From: Kris Kennaway To: Khairil Yusof Cc: Kris Kennaway , freebsd-smp@FreeBSD.ORG Subject: Re: Temporary lockups on 5-release Message-ID: <20030304203353.GB93052@rot13.obsecurity.org> References: <1046755156.598.92.camel@daemon.home.net> <20030304074800.GA89765@rot13.obsecurity.org> <1046786866.578.197.camel@daemon.home.net> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="KFztAG8eRSV9hGtP" Content-Disposition: inline In-Reply-To: <1046786866.578.197.camel@daemon.home.net> User-Agent: Mutt/1.4i 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 --KFztAG8eRSV9hGtP Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Mar 04, 2003 at 10:07:48PM +0800, Khairil Yusof wrote: > These lockups are worrying for me because, they lead to system reboots. > And they are my personal benchmark for stability (ie can FreeBSD recover > or not when it happens). >=20 > FreeBSD 5 RC-1 -> system reboots > FreeBSD 5 RC-2 -> ok > FREEBSD 5 RC-3 and RELEASE -> ok > FREEBSD CURRENT (as of last weekend) -> system reboot. It sounds like you're seeing kernel panics which are going unnoticed (well, except for the reboot ;) because you don't have the system configured for crashdumps or serial gdb. > I'm back to RELEASE, because the constant reboots can cause data > corruption which is not good for my one and only home system. That sounds appropriate. Kris --KFztAG8eRSV9hGtP Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (FreeBSD) iD8DBQE+ZQ2wWry0BWjoQKURAtEFAKC3zWoPyL77mG7H01anfRxOrch7CQCg8HHi QDtdPF6RO0IVvwrKd0xkNlY= =aer2 -----END PGP SIGNATURE----- --KFztAG8eRSV9hGtP-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 5 0:53:35 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 6CA9137B401 for ; Wed, 5 Mar 2003 00:53:30 -0800 (PST) Received: from mx0.gmx.net (mx0.gmx.net [213.165.64.100]) by mx1.FreeBSD.org (Postfix) with SMTP id EE0B043FBF for ; Wed, 5 Mar 2003 00:53:28 -0800 (PST) (envelope-from pcc@gmx.net) Received: (qmail 14452 invoked by uid 0); 5 Mar 2003 08:53:28 -0000 Date: Wed, 5 Mar 2003 09:53:27 +0100 (MET) From: Peter Cornelius To: smp@freebsd.org Cc: Julian Elischer MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="========GMXBoundary290111046854407" Subject: Re: UP ans SMP on the same kernel X-Priority: 3 (Normal) X-Authenticated-Sender: #0000491680@gmx.net X-Authenticated-IP: [217.83.5.114] Message-ID: <29011.1046854407@www62.gmx.net> X-Mailer: WWW-Mail 1.6 (Global Message Exchange) X-Flags: 0001 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 This is a MIME encapsulated multipart message - please use a MIME-compliant e-mail program to open it. Dies ist eine mehrteilige Nachricht im MIME-Format - bitte verwenden Sie zum Lesen ein MIME-konformes Mailprogramm. --========GMXBoundary290111046854407 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit References: <3821.1046762486@www54.gmx.net> <16352.1046763632@www54.gmx.net> In-Reply-To: <16352.1046763632@www54.gmx.net> On Tuesday 04 March 2003 07:40, Peter Cornelius wrote: > It really is a cheapo all-in-there-complete box from the non-food heap of a > food chain supermarket which I did not even get a > proper main board manual for but I have been made to believe it's a > MicroStar? MS-6399 (???) or something like that (I'd have to > check when I'm back home if that is of interest). However, the board > manufacturer only point to the distributor and they only point > to their press-this-key-here-to-start-windows 'handbook'. (Sigh) Just to confirm, it's a Medion MD 3000, apparently made by MSI as OEM model MS-6399. No (proper) docs available anywhere I know. Except from the connector positions on the left hand edge and very bottom left corner, and from what one can guess from that image, the board looks very much like the one shown here: http://www.watch.impress.co.jp/pc/docs/article/20010605/mb_msi2.jpg dmesg.boot somehow is lost during boot (rare/weird/crook diskless setup on my side?) but, for the curious among you, I'll attach an output of mptable -verbose and the active, i. e. uncommented lines of my kernel config file (it's a bit too messy with all lines in ;-)). 'hope this helps... Best regards, Peter. -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte l踄heln! Fotogalerie online mit GMX ohne eigene Homepage! --========GMXBoundary290111046854407 Content-Type: text/plain; name="mptable-verbose.20030305" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="mptable-verbose.20030305" Cj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT0KCk1QVGFibGUsIHZlcnNpb24gMi4wLjE1CgogbG9va2lu ZyBmb3IgRUJEQSBwb2ludGVyIEAgMHgwNDBlLCBmb3VuZCwgc2VhcmNoaW5nIEVCREEgQCAweDAw MDAyMGEwCiBzZWFyY2hpbmcgQ01PUyAndG9wIG9mIG1lbScgQCAweDAwMDlmYzAwICg2MzlLKQog c2VhcmNoaW5nIEJJT1MgQCAweDAwMGYwMDAwCgogTVAgRlBTIGZvdW5kIGluIEJJT1MgQCBwaHlz aWNhbCBhZGRyOiAweDAwMGY1YzEwCgotLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tCgpNUCBGbG9hdGlu ZyBQb2ludGVyIFN0cnVjdHVyZToKCiAgbG9jYXRpb246CQkJQklPUwogIHBoeXNpY2FsIGFkZHJl c3M6CQkweDAwMGY1YzEwCiAgc2lnbmF0dXJlOgkJCSdfTVBfJwogIGxlbmd0aDoJCQkxNiBieXRl cwogIHZlcnNpb246CQkJMS4xCiAgY2hlY2tzdW06CQkJMHg4MAogIG1vZGU6CQkJCVZpcnR1YWwg V2lyZQoKLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLQoKTVAgQ29uZmlnIFRhYmxlIEhlYWRlcjoKCiAg cGh5c2ljYWwgYWRkcmVzczoJCTB4MDAwZjE0MDAKICBzaWduYXR1cmU6CQkJJ1BDTVAnCiAgYmFz ZSB0YWJsZSBsZW5ndGg6CQkzMzYKICB2ZXJzaW9uOgkJCTEuMQogIGNoZWNrc3VtOgkJCTB4N2EK ICBPRU0gSUQ6CQkJJ09FTTAwMDAwJwogIFByb2R1Y3QgSUQ6CQkJJ1BST0QwMDAwMDAwMCcKICBP RU0gdGFibGUgcG9pbnRlcjoJCTB4MDAwMDAwMDAKICBPRU0gdGFibGUgc2l6ZToJCTAKICBlbnRy eSBjb3VudDoJCQkzNQogIGxvY2FsIEFQSUMgYWRkcmVzczoJCTB4ZmVlMDAwMDAKICBleHRlbmRl ZCB0YWJsZSBsZW5ndGg6CTAKICBleHRlbmRlZCB0YWJsZSBjaGVja3N1bToJMAoKLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLQoKTVAgQ29uZmlnIEJhc2UgVGFibGUgRW50cmllczoKCi0tClByb2Nlc3Nv cnM6CUFQSUMgSUQJVmVyc2lvbglTdGF0ZQkJRmFtaWx5CU1vZGVsCVN0ZXAJRmxhZ3MKCQkgMAkg MHgxMQkgQlNQLCB1c2FibGUJIDE1CSAxCSAyCSAweGZiZmYKLS0KQnVzOgkJQnVzIElECVR5cGUK CQkgMAkgUENJICAgCgkJIDEJIFBDSSAgIAoJCSAyCSBQQ0kgICAKCQkgMwkgSVNBICAgCi0tCkkv TyBBUElDczoJQVBJQyBJRAlWZXJzaW9uCVN0YXRlCQlBZGRyZXNzCgkJIDIJIDB4MTEJIHVzYWJs ZQkJIDB4ZmVjMDAwMDAKLS0KSS9PIEludHM6CVR5cGUJUG9sYXJpdHkgICAgVHJpZ2dlcglCdXMg SUQJIElSUQlBUElDIElECVBJTiMKCQlJTlQJYWN0aXZlLWxvICAgICAgIGxldmVsCSAgICAgMAkz MTpCCSAgICAgIDIJICAxNwoJCUlOVAlhY3RpdmUtbG8gICAgICAgbGV2ZWwJICAgICAwCTMxOkQJ ICAgICAgMgkgIDE5CgkJSU5UCWFjdGl2ZS1sbyAgICAgICBsZXZlbAkgICAgIDAJMzE6QwkgICAg ICAyCSAgMjMKCQlJTlQJYWN0aXZlLWxvICAgICAgIGxldmVsCSAgICAgMgkgMjpBCSAgICAgIDIJ ICAxOAoJCUlOVAlhY3RpdmUtbG8gICAgICAgbGV2ZWwJICAgICAyCSAzOkEJICAgICAgMgkgIDIy CgkJSU5UCWFjdGl2ZS1sbyAgICAgICBsZXZlbAkgICAgIDIJIDE6QQkgICAgICAyCSAgMTcKCQlJ TlQJYWN0aXZlLWxvICAgICAgIGxldmVsCSAgICAgMgkgNzpBCSAgICAgIDIJICAyMAoJCUlOVAlh Y3RpdmUtbG8gICAgICAgbGV2ZWwJICAgICAxCSAwOkEJICAgICAgMgkgIDE2CgkJSU5UCWFjdGl2 ZS1sbyAgICAgICBsZXZlbAkgICAgIDIJIDA6QQkgICAgICAyCSAgMTYKCQlJTlQJYWN0aXZlLWxv ICAgICAgIGxldmVsCSAgICAgMgkgNjpBCSAgICAgIDIJICAyMQoJCUlOVAlhY3RpdmUtbG8gICAg ICAgbGV2ZWwJICAgICAwCTMxOkEJICAgICAgMgkgIDE2CgkJRXh0SU5UCSBjb25mb3JtcyAgICBj b25mb3JtcwkgICAgIDMJICAgMAkgICAgICAyCSAgIDAKCQlJTlQJIGNvbmZvcm1zICAgIGNvbmZv cm1zCSAgICAgMwkgICAxCSAgICAgIDIJICAgMQoJCUlOVAkgY29uZm9ybXMgICAgY29uZm9ybXMJ ICAgICAzCSAgIDAJICAgICAgMgkgICAyCgkJSU5UCSBjb25mb3JtcyAgICBjb25mb3JtcwkgICAg IDMJICAgMwkgICAgICAyCSAgIDMKCQlJTlQJIGNvbmZvcm1zICAgIGNvbmZvcm1zCSAgICAgMwkg ICA0CSAgICAgIDIJICAgNAoJCUlOVAkgY29uZm9ybXMgICAgY29uZm9ybXMJICAgICAzCSAgIDUJ ICAgICAgMgkgICA1CgkJSU5UCSBjb25mb3JtcyAgICBjb25mb3JtcwkgICAgIDMJICAgNgkgICAg ICAyCSAgIDYKCQlJTlQJIGNvbmZvcm1zICAgIGNvbmZvcm1zCSAgICAgMwkgICA3CSAgICAgIDIJ ICAgNwoJCUlOVAlhY3RpdmUtaGkgICAgICAgIGVkZ2UJICAgICAzCSAgIDgJICAgICAgMgkgICA4 CgkJSU5UCSBjb25mb3JtcyAgICBjb25mb3JtcwkgICAgIDMJICAgOQkgICAgICAyCSAgIDkKCQlJ TlQJIGNvbmZvcm1zICAgIGNvbmZvcm1zCSAgICAgMwkgIDEwCSAgICAgIDIJICAxMAoJCUlOVAkg Y29uZm9ybXMgICAgY29uZm9ybXMJICAgICAzCSAgMTEJICAgICAgMgkgIDExCgkJSU5UCSBjb25m b3JtcyAgICBjb25mb3JtcwkgICAgIDMJICAxMgkgICAgICAyCSAgMTIKCQlJTlQJIGNvbmZvcm1z ICAgIGNvbmZvcm1zCSAgICAgMwkgIDEzCSAgICAgIDIJICAxMwoJCUlOVAkgY29uZm9ybXMgICAg Y29uZm9ybXMJICAgICAzCSAgMTQJICAgICAgMgkgIDE0CgkJSU5UCSBjb25mb3JtcyAgICBjb25m b3JtcwkgICAgIDMJICAxNQkgICAgICAyCSAgMTUKLS0KTG9jYWwgSW50czoJVHlwZQlQb2xhcml0 eSAgICBUcmlnZ2VyCUJ1cyBJRAkgSVJRCUFQSUMgSUQJUElOIwoJCUV4dElOVAkgY29uZm9ybXMg ICAgY29uZm9ybXMJICAgICAzCSAgIDAJICAgIDI1NQkgICAwCgkJTk1JCSBjb25mb3JtcyAgICBj b25mb3JtcwkgICAgIDMJICAgMAkgICAgMjU1CSAgIDEKCj09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0K Cg== --========GMXBoundary290111046854407 Content-Type: text/plain; name="PETRA.active" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="PETRA.active" bWFjaGluZQkJaTM4NgppZGVudAkJUEVUUkEKbWF4dXNlcnMJMApvcHRpb25zIAlNQVhEU0laPSIo NTEyKjEwMjQqMTAyNCkiCm9wdGlvbnMgCU1BWFNTSVo9Iig1MTIqMTAyNCoxMDI0KSIKb3B0aW9u cyAJREZMRFNJWj0iKDEyOCoxMDI0KjEwMjQpIgpvcHRpb25zIAlQUV9DQUNIRVNJWkU9NTEyCSMg Y29sb3IgZm9yIDUxMmsvMTZrIGNhY2hlCm9wdGlvbnMgCUlOQ0xVREVfQ09ORklHX0ZJTEUgICAg ICMgSW5jbHVkZSB0aGlzIGZpbGUgaW4ga2VybmVsCm9wdGlvbnMgCVNNUAkJCSMgU3ltbWV0cmlj IE11bHRpUHJvY2Vzc29yIEtlcm5lbApvcHRpb25zCQlDUFVfRU5BQkxFX1NTRQpvcHRpb25zIAlB UElDX0lPCQkJIyBTeW1tZXRyaWMgKEFQSUMpIEkvTwpjcHUJCUk2ODZfQ1BVCQkjIGFrYSBQZW50 aXVtIFBybyh0bSkKb3B0aW9ucwkJTUFUSF9FTVVMQVRFCSMgU2VlIHdoZXRoZXIgdGhpcyByZW1v dmVzIHRoZSBuZXh1cyBib290IHByb2IKb3B0aW9ucyAJQ09NUEFUXzQzCm9wdGlvbnMgCVVTRVJf TERUCQkjYWxsb3cgdXNlci1sZXZlbCBjb250cm9sIG9mIGkzODYgbGR0Cm9wdGlvbnMgCVNZU1ZT SE0JCSMgaW5jbHVkZSBzdXBwb3J0IGZvciBzaGFyZWQgbWVtb3J5Cm9wdGlvbnMgCVNZU1ZTRU0J CSMgaW5jbHVkZSBzdXBwb3J0IGZvciBzZW1hcGhvcmVzCm9wdGlvbnMgCVNZU1ZNU0cJCSMgaW5j bHVkZSBzdXBwb3J0IGZvciBtZXNzYWdlIHF1ZXVlcwpvcHRpb25zIAlEREIKb3B0aW9ucyAJRERC X1VOQVRURU5ERUQKb3B0aW9ucyAJS1RSQUNFCQkJI2tlcm5lbCB0cmFjaW5nCm9wdGlvbnMgCVBF UkZNT04Kb3B0aW9ucyAJVUNPTlNPTEUKb3B0aW9ucyAJVVNFUkNPTkZJRwkJI2Jvb3QgLWMgZWRp dG9yCm9wdGlvbnMgCVZJU1VBTF9VU0VSQ09ORklHCSN2aXN1YWwgYm9vdCAtYyBlZGl0b3IKb3B0 aW9ucyAJSU5FVAkJCSNJbnRlcm5ldCBjb21tdW5pY2F0aW9ucyBwcm90b2NvbHMKb3B0aW9ucyAJ SU5FVDYJCQkjSVB2NiBjb21tdW5pY2F0aW9ucyBwcm90b2NvbHMKb3B0aW9ucyAJSVBTRUMJCQkj SVAgc2VjdXJpdHkKb3B0aW9ucyAJSVBTRUNfRVNQCQkjSVAgc2VjdXJpdHkgKGNyeXB0bzsgZGVm aW5lIHcvIElQU0VDKQpvcHRpb25zIAlJUFgJCQkjSVBYL1NQWCBjb21tdW5pY2F0aW9ucyBwcm90 b2NvbHMKb3B0aW9ucyAJTkVUR1JBUEgJCSNuZXRncmFwaCg0KSBzeXN0ZW0KcHNldWRvLWRldmlj ZQlldGhlcgkJCSNHZW5lcmljIEV0aGVybmV0CnBzZXVkby1kZXZpY2UJbG9vcAkJCSNOZXR3b3Jr IGxvb3BiYWNrIGRldmljZQpwc2V1ZG8tZGV2aWNlCWJwZgkJCSNCZXJrZWxleSBwYWNrZXQgZmls dGVyCnBzZXVkby1kZXZpY2UJZGlzYwkJCSNEaXNjYXJkIGRldmljZSAoZHMwLCBkczEsIGV0YykK cHNldWRvLWRldmljZQl0dW4JMgkJI1R1bm5lbCBkcml2ZXIgKHBwcCg4KSwgbm9zLXR1big4KSkK cHNldWRvLWRldmljZQlwcHAJMgkJI1BvaW50LXRvLXBvaW50IHByb3RvY29sCm9wdGlvbnMgCVBQ UF9CU0RDT01QCQkjUFBQIEJTRC1jb21wcmVzcyBzdXBwb3J0Cm9wdGlvbnMgCVBQUF9ERUZMQVRF CQkjUFBQIHpsaWIvZGVmbGF0ZS9nemlwIHN1cHBvcnQKb3B0aW9ucyAJUFBQX0ZJTFRFUgkJI2Vu YWJsZSBicGYgZmlsdGVyaW5nIChuZWVkcyBicGYpCnBzZXVkby1kZXZpY2UJZ2lmCQkJI0lQdjYg YW5kIElQdjQgdHVubmVsaW5nCnBzZXVkby1kZXZpY2UJZmFpdGgJMQkJI2ZvciBJUHY2IGFuZCBJ UHY0IHRyYW5zbGF0aW9uCnBzZXVkby1kZXZpY2UJc3RmCQkJIzZ0bzQgSVB2NiBvdmVyIElQdjQg ZW5jYXBzdWxhdGlvbgpvcHRpb25zIAlNUk9VVElORwkJIyBNdWx0aWNhc3Qgcm91dGluZwpvcHRp b25zIAlSQU5ET01fSVBfSUQKb3B0aW9ucyAJSUNNUF9CQU5ETElNCm9wdGlvbnMgCUZGUwkJCSNG YXN0IGZpbGVzeXN0ZW0Kb3B0aW9ucyAJTUZTCQkJI01lbW9yeSBGaWxlIFN5c3RlbQpvcHRpb25z IAlORlMJCQkjTmV0d29yayBGaWxlIFN5c3RlbQpvcHRpb25zIAlGRlNfUk9PVAkJI0ZGUyB1c2Fi bGUgYXMgcm9vdCBkZXZpY2UKb3B0aW9ucyAJTkZTX1JPT1QJCSNORlMgdXNhYmxlIGFzIHJvb3Qg ZGV2aWNlCm9wdGlvbnMgCVNPRlRVUERBVEVTCm9wdGlvbnMgCVVGU19ESVJIQVNICm9wdGlvbnMg CU1EX1JPT1QKb3B0aW9ucyAJUVVPVEEJCQkjZW5hYmxlIGRpc2sgcXVvdGFzCm9wdGlvbnMgCVAx MDAzXzFCCm9wdGlvbnMgCV9LUE9TSVhfUFJJT1JJVFlfU0NIRURVTElORwpvcHRpb25zIAlfS1BP U0lYX1ZFUlNJT049MTk5MzA5TApwc2V1ZG8tZGV2aWNlCXB0eQkJI1BzZXVkbyB0dHlzCnBzZXVk by1kZXZpY2UJc3BlYWtlcgkJI1BsYXkgSUJNIEJBU0lDLXN0eWxlIG5vaXNlcyBvdXQgeW91ciBz cGVha2VyCnBzZXVkby1kZXZpY2UJZ3ppcAkJI0V4ZWMgZ3ppcHBlZCBhLm91dCdzCnBzZXVkby1k ZXZpY2UJdm4JCSNWbm9kZSBkcml2ZXIgKHR1cm5zIGEgZmlsZSBpbnRvIGEgZGV2aWNlKQpwc2V1 ZG8tZGV2aWNlCW1kCQkjTWVtb3J5L21hbGxvYyBkaXNrCnBzZXVkby1kZXZpY2UJc25wCQkjU25v b3AgZGV2aWNlIC0gdG8gbG9vayBhdCBwdHkvdnR5L2V0Yy4uCmRldmljZQkJaXNhCm9wdGlvbnMg CUFVVE9fRU9JXzEKb3B0aW9ucyAJQVVUT19FT0lfMgpkZXZpY2UJCWF0a2JkYzAJYXQgaXNhPyBw b3J0IElPX0tCRApkZXZpY2UJCWF0a2JkMAlhdCBhdGtiZGM/IGlycSAxCm9wdGlvbnMgCUtCRF9J TlNUQUxMX0NERVYJIyBpbnN0YWxsIGEgQ0RFViBlbnRyeSBpbiAvZGV2CmRldmljZQkJcHNtMAlh dCBhdGtiZGM/IGlycSAxMgpkZXZpY2UJCXZnYTAJYXQgaXNhPwpvcHRpb25zIAlWR0FfV0lEVEg5 MAkJIyBzdXBwb3J0IDkwIGNvbHVtbiBtb2RlcwpvcHRpb25zIAlWRVNBCnBzZXVkby1kZXZpY2UJ c3BsYXNoCmRldmljZQkJc2MwCWF0IGlzYT8Kb3B0aW9ucyAJTUFYQ09OUz0xMgkJIyBudW1iZXIg b2YgdmlydHVhbCBjb25zb2xlcwpvcHRpb25zIAlTQ19ISVNUT1JZX1NJWkU9NDAwCSMgbnVtYmVy IG9mIGhpc3RvcnkgYnVmZmVyIGxpbmVzCm9wdGlvbnMgCVNDX01PVVNFX0NIQVI9MHgzCSMgY2hh ciBjb2RlIGZvciB0ZXh0IG1vZGUgbW91c2UgY3Vyc29yCm9wdGlvbnMgCVNDX1BJWEVMX01PREUJ CSMgYWRkIHN1cHBvcnQgZm9yIHRoZSByYXN0ZXIgdGV4dCBtb2RlCm9wdGlvbnMgCVNDX05PUk1f QVRUUj0iKEZHX0dSRUVOfEJHX0JMQUNLKSIKb3B0aW9ucyAJU0NfTk9STV9SRVZfQVRUUj0iKEZH X1lFTExPV3xCR19HUkVFTikiCm9wdGlvbnMgCVNDX0tFUk5FTF9DT05TX0FUVFI9IihGR19SRUR8 QkdfQkxBQ0spIgpvcHRpb25zIAlTQ19LRVJORUxfQ09OU19SRVZfQVRUUj0iKEZHX0JMQUNLfEJH X1JFRCkiCmRldmljZQkJbnB4MAlhdCBuZXh1cz8gcG9ydCBJT19OUFggaXJxIDEzCmRldmljZQkJ YXRhCmRldmljZQkJYXRhZGlzawkJIyBBVEEgZGlzayBkcml2ZXMKZGV2aWNlCQlhdGFwaWNkCQkj IEFUQVBJIENEUk9NIGRyaXZlcwpkZXZpY2UJCWF0YXBpZmQJCSMgQVRBUEkgZmxvcHB5IGRyaXZl cwpkZXZpY2UJCWF0YXBpc3QJCSMgQVRBUEkgdGFwZSBkcml2ZXMKb3B0aW9ucyAJQVRBX1NUQVRJ Q19JRApkZXZpY2UJCWZkYzAJYXQgaXNhPyBwb3J0IElPX0ZEMSBpcnEgNiBkcnEgMgpkZXZpY2UJ CWZkMAlhdCBmZGMwIGRyaXZlIDAKZGV2aWNlCQlmZDEJYXQgZmRjMCBkcml2ZSAxCmRldmljZQkJ c2lvMAlhdCBpc2E/IHBvcnQgSU9fQ09NMSBmbGFncyAweDIwMDEwIGlycSA0CmRldmljZQkJc2lv MQlhdCBpc2E/IHBvcnQgSU9fQ09NMiBmbGFncyAweDIwMDEwIGlycSAzCmRldmljZQkJc2lvMglh dCBpc2E/IHBvcnQgSU9fQ09NMyBkaXNhYmxlIGZsYWdzIDB4MjAwMTAgaXJxIDUKZGV2aWNlCQlz aW8zCWF0IGlzYT8gcG9ydCBJT19DT000IGRpc2FibGUgZmxhZ3MgMHgyMDAxMCBpcnEgOQpkZXZp Y2UJCXBjbQpkZXZpY2UJCXBjYTAgYXQgaXNhPyBwb3J0IElPX1RJTUVSMQpkZXZpY2UJCWFwbTAK ZGV2aWNlCQllaXNhCmRldmljZQkJcGNpCmRldmljZSBlZDAgYXQgcGNpPyAJIyBlZDogV2VzdGVy biBEaWdpdGFsIGFuZCBTTUMgODB4eDsgTm92ZWxsIE5FMTAwMCBhbmQgTkUyMDAwOyAzQ29tIDND NTAzCmRldmljZQkJYWdwCmRldmljZQkJbWlpYnVzCmRldmljZQkJcmwJCSMgUmVhbFRlayA4MTI5 LzgxMzkKZGV2aWNlCQlia3RyCmRldmljZQkJc21idXMJCSMgQnVzIHN1cHBvcnQsIHJlcXVpcmVk IGZvciBzbWIgYmVsb3cuCmRldmljZQkJaW50cG0KZGV2aWNlCQlhbHBtCmRldmljZQkJaWNoc21i CmRldmljZQkJdmlhcG0KZGV2aWNlCQlhbWRwbQpkZXZpY2UJCXNtYgpkZXZpY2UJCWlpY2J1cwkJ IyBCdXMgc3VwcG9ydCwgcmVxdWlyZWQgZm9yIGljL2lpYy9paWNzbWIgYmVsb3cuCmRldmljZQkJ aWljYmIKZGV2aWNlCQlpYwpkZXZpY2UJCWlpYwpkZXZpY2UJCWlpY3NtYgkJIyBzbWIgb3ZlciBp MmMgYnJpZGdlCm9wdGlvbnMgCVBQQ19QUk9CRV9DSElQU0VUICMgRW5hYmxlIGNoaXBzZXQgc3Bl Y2lmaWMgZGV0ZWN0aW9uCmRldmljZQkJcHBjMAlhdCBpc2E/IGlycSA3CmRldmljZQkJcHBidXMK ZGV2aWNlCQlscHQKZGV2aWNlCQlwbGlwCmRldmljZQkJcHBpCmRldmljZQkJbHBiYgpvcHRpb25z IAlCT09UUAkJIyBVc2UgQk9PVFAgdG8gb2J0YWluIElQIGFkZHJlc3MvaG9zdG5hbWUKb3B0aW9u cyAJQk9PVFBfTkZTUk9PVAkjIE5GUyBtb3VudCByb290IGZpbGVzeXN0ZW0gdXNpbmcgQk9PVFAg aW5mbwpvcHRpb25zIAlCT09UUF9ORlNWMwkjIFVzZSBORlMgdjMgdG8gTkZTIG1vdW50IHJvb3QK ZGV2aWNlCQl1aGNpCmRldmljZQkJb2hjaQpkZXZpY2UJCXVzYgpkZXZpY2UJCXVnZW4KZGV2aWNl CQl1aGlkCmRldmljZQkJZmlyZXdpcmUJIyBGaXJld2lyZSBidXMgY29kZQpvcHRpb25zIAlQTlBC SU9TCm9wdGlvbnMgCVNIT1dfQlVTWUJVRlMJIyBMaXN0IGJ1ZmZlcnMgdGhhdCBwcmV2ZW50IHJv b3QgdW5tb3VudAo= --========GMXBoundary290111046854407-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Mar 6 3:10:30 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9B00137B401 for ; Thu, 6 Mar 2003 03:10:28 -0800 (PST) Received: from c-24-130-253-104.we.client2.attbi.com (c-24-130-253-104.we.client2.attbi.com [24.130.253.104]) by mx1.FreeBSD.org (Postfix) with ESMTP id 18F9843F85 for ; Thu, 6 Mar 2003 03:10:28 -0800 (PST) (envelope-from obrien@c-24-130-253-104.we.client2.attbi.com) Received: from localhost (localhost [127.0.0.1]) by c-24-130-253-104.we.client2.attbi.com (8.11.1/8.11.1) with ESMTP id h26B9nd00841 for ; Thu, 6 Mar 2003 03:09:49 -0800 (PST) (envelope-from obrien@c-24-130-253-104.we.client2.attbi.com) Message-Id: <200303061109.h26B9nd00841@c-24-130-253-104.we.client2.attbi.com> X-Mailer: exmh version 2.2 06/23/2000 with version: MH 6.8.4 #1[UCI] To: freebsd-smp@freebsd.org Subject: SMP hang on boot in RELENG_4_7 on Iwill DP533 board Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 06 Mar 2003 03:09:48 -0800 From: "Mike O'Brien" 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'm running dual Xeon processors on an Iwill DP533 board. This board has an E7505 MCH chipset from Intel, using the ICH4 I/O hub and an 82870P2 PCI-X controller. This configuration does the same thing that John Stockdale reported on his Tyan Thunder i7505 board: it runs fine with a uniprocessor kernel, but attempting to boot an SMP kernel results in a hang at the point where the kernel has just printed "APIC_IO: Testing 8254 interrupt delivery." Rather than muck with my BIOS as he did, I mucked with the kernel. This message is printed by clock.c, just before executing a "while" loop waiting for some I/O register action. That action never happens. Not figuring it would work, but figuring it'd at least be instructive, I inserted a count-backward-from-a-million escape hatch so the loop would run for a while, then exit. The resulting kernel did boot, and did turn on the second processor, and seemed to run OK until I tried to do a "make installworld". I got a ton of disk read and write timeout errors, probably because the APIC_IO wasn't, in fact, set up correctly. My understanding is that a good part of these problems are that the developers don't have access to the hardware needed to test stuff out. Any SMP developer who wants to work with me on testing fixes to this problem, send me private email, please, and let's see if we can work something out. I for dang sure don't want to be limited to a single processor. Suggestions welcome. I don't know, for example, if the APIC setup in 5.0 has been improved to the point where I should try running it. I was avoiding it, but I could give it a try if there's a chance it'll work. Comments? Mike O'Brien To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Mar 6 10:21:36 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id DA22237B401 for ; Thu, 6 Mar 2003 10:21:33 -0800 (PST) Received: from smtp1.Stanford.EDU (smtp1.Stanford.EDU [171.64.14.23]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1E3D343F93 for ; Thu, 6 Mar 2003 10:21:33 -0800 (PST) (envelope-from jstockdale@stanford.edu) Received: from smtp1.Stanford.EDU ([127.0.0.1]) by smtp1.Stanford.EDU (8.12.6/8.12.6) with ESMTP id h26ILWHI007978 for ; Thu, 6 Mar 2003 10:21:32 -0800 (PST) Received: from quenya (quenya.Stanford.EDU [128.12.44.61]) by smtp1.Stanford.EDU (8.12.6/8.12.6) with ESMTP id h26ILTDS007963 for ; Thu, 6 Mar 2003 10:21:31 -0800 (PST) From: "John Stockdale" To: Subject: RE: SMP hang on boot in RELENG_4_7 on Iwill DP533 board Date: Thu, 6 Mar 2003 10:21:19 -0800 Message-ID: <002401c2e40d$303199e0$3d2c0c80@quenya> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.2627 Importance: Normal In-Reply-To: <200303061109.h26B9nd00841@c-24-130-253-104.we.client2.attbi.com> X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 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 was actually having the same problem with 4.7-RELEASE and 5.0-RELEASE, except I was using a Tyan i7505 board (same chipset, comparable board I believe). Anyway, to make a long story short, the problem I was having stemmed from a problem in the code that was merged after 4.7-RELEASE and didn't actually make it in until Early Feb. As soon as I installed 4.8-RC1 the SMP kernel booted fine and found all 4 processors (well 2 x 2 (HyperThreading)). Good Luck -John -----Original Message----- From: owner-freebsd-smp@FreeBSD.ORG [mailto:owner-freebsd-smp@FreeBSD.ORG] On Behalf Of Mike O'Brien Sent: Thursday, March 06, 2003 3:10 AM To: freebsd-smp@FreeBSD.ORG Subject: SMP hang on boot in RELENG_4_7 on Iwill DP533 board I'm running dual Xeon processors on an Iwill DP533 board. This board has an E7505 MCH chipset from Intel, using the ICH4 I/O hub and an 82870P2 PCI-X controller. This configuration does the same thing that John Stockdale reported on his Tyan Thunder i7505 board: it runs fine with a uniprocessor kernel, but attempting to boot an SMP kernel results in a hang at the point where the kernel has just printed "APIC_IO: Testing 8254 interrupt delivery." Rather than muck with my BIOS as he did, I mucked with the kernel. This message is printed by clock.c, just before executing a "while" loop waiting for some I/O register action. That action never happens. Not figuring it would work, but figuring it'd at least be instructive, I inserted a count-backward-from-a-million escape hatch so the loop would run for a while, then exit. The resulting kernel did boot, and did turn on the second processor, and seemed to run OK until I tried to do a "make installworld". I got a ton of disk read and write timeout errors, probably because the APIC_IO wasn't, in fact, set up correctly. My understanding is that a good part of these problems are that the developers don't have access to the hardware needed to test stuff out. Any SMP developer who wants to work with me on testing fixes to this problem, send me private email, please, and let's see if we can work something out. I for dang sure don't want to be limited to a single processor. Suggestions welcome. I don't know, for example, if the APIC setup in 5.0 has been improved to the point where I should try running it. I was avoiding it, but I could give it a try if there's a chance it'll work. Comments? Mike O'Brien To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Mar 6 11:45:49 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3B43A37B401 for ; Thu, 6 Mar 2003 11:45:47 -0800 (PST) Received: from chunky.global.net.uk (chunky.global.net.uk [80.189.91.76]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2A6C843F3F for ; Thu, 6 Mar 2003 11:45:46 -0800 (PST) (envelope-from john.ekins@brightview.com) Received: from ravms by chunky.global.net.uk with mail-ok (Exim 3.36 #1) id 18r1Ej-0005tr-00 for freebsd-smp@freebsd.org; Thu, 06 Mar 2003 19:41:01 +0000 Received: from [195.137.100.169] (helo=chunky.internal.amaretti.net) by chunky.global.net.uk with smtp (Exim 3.36 #1) id 18r1Ei-0005tl-00 for freebsd-smp@freebsd.org; Thu, 06 Mar 2003 19:41:00 +0000 Date: Thu, 6 Mar 2003 19:44:00 +0000 From: John Ekins To: freebsd-smp@freebsd.org Subject: Regular panic on SMP system Message-Id: <20030306194400.4fa939e3.john.ekins@brightview.com> Organization: Brightview X-Mailer: Sylpheed version 0.8.6 (GTK+ 1.2.10; i386-portbld-freebsd5.0) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Envelope-To: freebsd-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 Gents/ladies(!), I'm running several SMP systems but am having regular problems with one of them. I'm not sure if this is really a question for the SMP or the questions list, but I'm trying here first. The machines are dual Xeon 2.4GHz .5MB cache with hyperthreading enabled. They are less than 2 weeks old. They have identical kernels, which leads me to believe this is a hardware problem with this particular machine but I'm not too sure. I'm looking for some advice. I've taken the liberty of posting the log when the machine crashes below. I thought that "Fatal trap 12: page fault while in kernel mode" referred to below suggests that something has eaten up the stack. These machines run Exim + named + anti virus software. I consider them lightly loaded. Does anyone have any suggestions, and if I should post this to questions I apologise. Cheers, John. Mar 6 10:12:45 mx0 /kernel: Fatal trap 12: page fault while in kernel mode Mar 6 10:12:45 mx0 /kernel: mp_lock = 03000002; cpuid = 3; lapic.id = 03000000 Mar 6 10:12:45 mx0 /kernel: fault virtual address = 0x0 Mar 6 10:12:45 mx0 /kernel: fault code = supervisor write, page not present Mar 6 10:12:45 mx0 /kernel: instruction pointer = 0x8:0xc01995ca Mar 6 10:12:45 mx0 /kernel: stack pointer = 0x10:0xe4bdbecc Mar 6 10:12:45 mx0 /kernel: frame pointer = 0x10:0xe4bdbed0 Mar 6 10:12:45 mx0 /kernel: code segment = base 0x0, limit 0xfffff, type 0x1b Mar 6 10:12:45 mx0 /kernel: = DPL 0, pres 1, def32 1, gran 1 Mar 6 10:12:45 mx0 /kernel: processor eflags = interrupt enabled, resume, IOPL = 0 Mar 6 10:12:45 mx0 /kernel: current process = 6 (vnlru) Mar 6 10:12:45 mx0 /kernel: interrupt mask = bio <- SMP: XXX Mar 6 10:12:45 mx0 /kernel: trap number = 12 Mar 6 10:12:45 mx0 /kernel: panic: page fault Mar 6 10:12:45 mx0 /kernel: mp_lock = 03000002; cpuid = 3; lapic.id = 03000000 Mar 6 10:12:45 mx0 /kernel: boot() called on cpu#3 Mar 6 10:12:45 mx0 /kernel: Mar 6 10:12:45 mx0 /kernel: syncing disks... 84 3 3 3 3 3 3 3 34 Mar 6 10:12:45 mx0 /kernel: done Mar 6 10:12:45 mx0 /kernel: Uptime: 3h8m39s Mar 6 10:12:45 mx0 /kernel: Automatic reboot in 15 seconds - press a key on the console to abort Mar 6 10:12:45 mx0 /kernel: Rebooting... Mar 6 10:12:45 mx0 /kernel: cpu_reset called on cpu#3 Mar 6 10:12:45 mx0 /kernel: cpu_reset: Stopping other CPUs Mar 6 10:12:45 mx0 /kernel: cpu_reset: Restarting BSP Mar 6 10:12:45 mx0 /kernel: cpu_reset_proxy: Gcrpaub_breeds emtp: lBoScPk dfiodr nBoStP Mar 6 10:12:45 mx0 /kernel: grab mp lock To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Mar 6 11:53:38 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 81FB737B401 for ; Thu, 6 Mar 2003 11:53:34 -0800 (PST) Received: from mx1.freebsdsystems.com (gw.tor.freebsdsystems.com [216.138.230.94]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6A03043F3F for ; Thu, 6 Mar 2003 11:53:31 -0800 (PST) (envelope-from lnb@freebsdsystems.com) Received: (qmail 99105 invoked by uid 89); 6 Mar 2003 19:53:30 -0000 Received: from unknown (HELO noemie) (192.168.0.2) by mx1.freebsdsystems.com with SMTP; 6 Mar 2003 19:53:30 -0000 Message-ID: <006101c2e41a$58238280$0200a8c0@noemie> From: "Lanny Baron" To: "John Ekins" , References: <20030306194400.4fa939e3.john.ekins@brightview.com> Subject: Re: Regular panic on SMP system Date: Thu, 6 Mar 2003 14:55:31 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1106 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 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 like that.. 'regular' Send me mine with extra sugar :) What Server Board are you using? Lanny ----- Original Message ----- From: "John Ekins" To: Sent: Thursday, March 06, 2003 2:44 PM Subject: Regular panic on SMP system > Gents/ladies(!), > > I'm running several SMP systems but am having regular problems with one of > them. I'm not sure if this is really a question for the SMP or the questions > list, but I'm trying here first. > > The machines are dual Xeon 2.4GHz .5MB cache with hyperthreading enabled. They > are less than 2 weeks old. They have identical kernels, which leads me to > believe this is a hardware problem with this particular machine but I'm not > too sure. I'm looking for some advice. I've taken the liberty of posting the > log when the machine crashes below. > > I thought that "Fatal trap 12: page fault while in kernel mode" referred to > below suggests that something has eaten up the stack. These machines run Exim > + named + anti virus software. I consider them lightly loaded. > > Does anyone have any suggestions, and if I should post this to questions I > apologise. > > Cheers, > John. > > > Mar 6 10:12:45 mx0 /kernel: Fatal trap 12: page fault while in kernel mode > Mar 6 10:12:45 mx0 /kernel: mp_lock = 03000002; cpuid = 3; lapic.id = > 03000000 Mar 6 10:12:45 mx0 /kernel: fault virtual address = 0x0 > Mar 6 10:12:45 mx0 /kernel: fault code = supervisor write, page not > present Mar 6 10:12:45 mx0 /kernel: instruction pointer = > 0x8:0xc01995ca > Mar 6 10:12:45 mx0 /kernel: stack pointer = 0x10:0xe4bdbecc > Mar 6 10:12:45 mx0 /kernel: frame pointer = 0x10:0xe4bdbed0 > Mar 6 10:12:45 mx0 /kernel: code segment = base > 0x0, limit 0xfffff, type 0x1b > Mar 6 10:12:45 mx0 /kernel: = DPL 0, pres 1, def32 1, gran 1 > Mar 6 10:12:45 mx0 /kernel: processor eflags = interrupt enabled, resume, > IOPL = 0 > Mar 6 10:12:45 mx0 /kernel: current process = 6 (vnlru) > Mar 6 10:12:45 mx0 /kernel: interrupt mask = bio <- SMP: XXX > Mar 6 10:12:45 mx0 /kernel: trap number = 12 > Mar 6 10:12:45 mx0 /kernel: panic: page fault > Mar 6 10:12:45 mx0 /kernel: mp_lock = 03000002; cpuid = 3; lapic.id = > 03000000 > Mar 6 10:12:45 mx0 /kernel: boot() called on cpu#3 > Mar 6 10:12:45 mx0 /kernel: > Mar 6 10:12:45 mx0 /kernel: syncing disks... 84 3 3 3 3 3 3 3 34 > Mar 6 10:12:45 mx0 /kernel: done > Mar 6 10:12:45 mx0 /kernel: Uptime: 3h8m39s > Mar 6 10:12:45 mx0 /kernel: Automatic reboot in 15 seconds - press a key on > the console to abort > Mar 6 10:12:45 mx0 /kernel: Rebooting... > Mar 6 10:12:45 mx0 /kernel: cpu_reset called on cpu#3 > Mar 6 10:12:45 mx0 /kernel: cpu_reset: Stopping other CPUs > Mar 6 10:12:45 mx0 /kernel: cpu_reset: Restarting BSP > Mar 6 10:12:45 mx0 /kernel: cpu_reset_proxy: Gcrpaub_breeds emtp: lBoScPk > dfiodr nBoStP > Mar 6 10:12:45 mx0 /kernel: grab mp lock > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Mar 6 12: 2:34 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id DB5DE37B401 for ; Thu, 6 Mar 2003 12:02:32 -0800 (PST) Received: from chunky.global.net.uk (chunky.global.net.uk [80.189.91.76]) by mx1.FreeBSD.org (Postfix) with ESMTP id D0A1E43FD7 for ; Thu, 6 Mar 2003 12:02:31 -0800 (PST) (envelope-from john.ekins@brightview.com) Received: from ravms by chunky.global.net.uk with mail-ok (Exim 3.36 #1) id 18r1Ux-0006G5-00; Thu, 06 Mar 2003 19:57:47 +0000 Received: from [195.137.100.169] (helo=chunky.internal.amaretti.net) by chunky.global.net.uk with smtp (Exim 3.36 #1) id 18r1Uw-0006Fz-00; Thu, 06 Mar 2003 19:57:47 +0000 Date: Thu, 6 Mar 2003 20:00:46 +0000 From: John Ekins To: "Lanny Baron" Cc: freebsd-smp@freebsd.org Subject: Re: Regular panic on SMP system Message-Id: <20030306200046.07ebfaa2.john.ekins@brightview.com> In-Reply-To: <006101c2e41a$58238280$0200a8c0@noemie> References: <20030306194400.4fa939e3.john.ekins@brightview.com> <006101c2e41a$58238280$0200a8c0@noemie> Organization: Brightview X-Mailer: Sylpheed version 0.8.6 (GTK+ 1.2.10; i386-portbld-freebsd5.0) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Envelope-To: lnb@freebsdsystems.com, freebsd-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 Thu, 6 Mar 2003 14:55:31 -0500 "Lanny Baron" wrote: -> I like that.. 'regular' Send me mine with extra sugar :) OK, irregular panic! Most irregular in fact compared with any of my other boxen. -> What Server Board are you using? A good question without opening it up. Here's some info from mptable if that helps: MP Config Table Header: physical address: 0x000f63f0 signature: 'PCMP' base table length: 388 version: 1.4 checksum: 0x92 OEM ID: 'GIGABYTE' Product ID: '8IPPDR ' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 37 local APIC address: 0xfee00000 extended table length: 124 extended table checksum: 251 Would the whole output + dmesg be of more use? -> Lanny Cheers, John. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Mar 6 12:15:59 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0BE4A37B401 for ; Thu, 6 Mar 2003 12:15:57 -0800 (PST) Received: from mail.speakeasy.net (mail11.speakeasy.net [216.254.0.211]) by mx1.FreeBSD.org (Postfix) with ESMTP id 680DF43F93 for ; Thu, 6 Mar 2003 12:15:56 -0800 (PST) (envelope-from jhb@FreeBSD.org) Received: (qmail 23744 invoked from network); 6 Mar 2003 20:16:02 -0000 Received: from unknown (HELO server.baldwin.cx) ([216.27.160.63]) (envelope-sender ) by mail11.speakeasy.net (qmail-ldap-1.03) with DES-CBC3-SHA encrypted SMTP for ; 6 Mar 2003 20:16:02 -0000 Received: from laptop.baldwin.cx (gw1.twc.weather.com [216.133.140.1]) by server.baldwin.cx (8.12.6/8.12.6) with ESMTP id h26KCxhT052486; Thu, 6 Mar 2003 15:13:00 -0500 (EST) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.5.2 on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <20030306194400.4fa939e3.john.ekins@brightview.com> Date: Thu, 06 Mar 2003 15:16:05 -0500 (EST) From: John Baldwin To: John Ekins Subject: RE: Regular panic on SMP system Cc: freebsd-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 06-Mar-2003 John Ekins wrote: > Gents/ladies(!), > > I'm running several SMP systems but am having regular problems with one of > them. I'm not sure if this is really a question for the SMP or the questions > list, but I'm trying here first. > > The machines are dual Xeon 2.4GHz .5MB cache with hyperthreading enabled. They > are less than 2 weeks old. They have identical kernels, which leads me to > believe this is a hardware problem with this particular machine but I'm not > too sure. I'm looking for some advice. I've taken the liberty of posting the > log when the machine crashes below. > > I thought that "Fatal trap 12: page fault while in kernel mode" referred to > below suggests that something has eaten up the stack. These machines run Exim > + named + anti virus software. I consider them lightly loaded. > > Does anyone have any suggestions, and if I should post this to questions I > apologise. > > Cheers, > John. > > > Mar 6 10:12:45 mx0 /kernel: Fatal trap 12: page fault while in kernel mode > Mar 6 10:12:45 mx0 /kernel: mp_lock = 03000002; cpuid = 3; lapic.id = > 03000000 Mar 6 10:12:45 mx0 /kernel: fault virtual address = 0x0 NULL pointer dereference. Any chance you could get a dump or get a stack trace from ddb maybe? > Mar 6 10:12:45 mx0 /kernel: fault code = supervisor write, page not > present Mar 6 10:12:45 mx0 /kernel: instruction pointer = > 0x8:0xc01995ca > Mar 6 10:12:45 mx0 /kernel: stack pointer = 0x10:0xe4bdbecc > Mar 6 10:12:45 mx0 /kernel: frame pointer = 0x10:0xe4bdbed0 > Mar 6 10:12:45 mx0 /kernel: code segment = base > 0x0, limit 0xfffff, type 0x1b > Mar 6 10:12:45 mx0 /kernel: = DPL 0, pres 1, def32 1, gran 1 > Mar 6 10:12:45 mx0 /kernel: processor eflags = interrupt enabled, resume, > IOPL = 0 > Mar 6 10:12:45 mx0 /kernel: current process = 6 (vnlru) > Mar 6 10:12:45 mx0 /kernel: interrupt mask = bio <- SMP: XXX > Mar 6 10:12:45 mx0 /kernel: trap number = 12 > Mar 6 10:12:45 mx0 /kernel: panic: page fault > Mar 6 10:12:45 mx0 /kernel: mp_lock = 03000002; cpuid = 3; lapic.id = > 03000000 > Mar 6 10:12:45 mx0 /kernel: boot() called on cpu#3 > Mar 6 10:12:45 mx0 /kernel: > Mar 6 10:12:45 mx0 /kernel: syncing disks... 84 3 3 3 3 3 3 3 34 > Mar 6 10:12:45 mx0 /kernel: done > Mar 6 10:12:45 mx0 /kernel: Uptime: 3h8m39s > Mar 6 10:12:45 mx0 /kernel: Automatic reboot in 15 seconds - press a key on > the console to abort > Mar 6 10:12:45 mx0 /kernel: Rebooting... > Mar 6 10:12:45 mx0 /kernel: cpu_reset called on cpu#3 > Mar 6 10:12:45 mx0 /kernel: cpu_reset: Stopping other CPUs > Mar 6 10:12:45 mx0 /kernel: cpu_reset: Restarting BSP > Mar 6 10:12:45 mx0 /kernel: cpu_reset_proxy: Gcrpaub_breeds emtp: lBoScPk > dfiodr nBoStP > Mar 6 10:12:45 mx0 /kernel: grab mp lock > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message -- John Baldwin <>< http://www.FreeBSD.org/~jhb/ "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 Thu Mar 6 12:26:43 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id CC78737B401; Thu, 6 Mar 2003 12:26:42 -0800 (PST) Received: from chunky.global.net.uk (chunky.global.net.uk [80.189.91.76]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3A86C43F3F; Thu, 6 Mar 2003 12:26:42 -0800 (PST) (envelope-from john.ekins@brightview.com) Received: from ravms by chunky.global.net.uk with mail-ok (Exim 3.36 #1) id 18r1sL-0006ee-00; Thu, 06 Mar 2003 20:21:57 +0000 Received: from [195.137.100.169] (helo=chunky.internal.amaretti.net) by chunky.global.net.uk with smtp (Exim 3.36 #1) id 18r1sL-0006eY-00; Thu, 06 Mar 2003 20:21:57 +0000 Date: Thu, 6 Mar 2003 20:24:56 +0000 From: John Ekins To: John Baldwin Cc: freebsd-smp@freebsd.org Subject: Re: Regular panic on SMP system Message-Id: <20030306202456.718e1898.john.ekins@brightview.com> In-Reply-To: References: <20030306194400.4fa939e3.john.ekins@brightview.com> Organization: Brightview X-Mailer: Sylpheed version 0.8.6 (GTK+ 1.2.10; i386-portbld-freebsd5.0) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Envelope-To: jhb@FreeBSD.org, freebsd-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 Thu, 06 Mar 2003 15:16:05 -0500 (EST) John Baldwin wrote: -> NULL pointer dereference. Any chance you could get a dump or get a stack -> trace from ddb maybe? I'm just building a kernel with DDB for this very purpose and have enabled savecore. I'll let you know if/when the machine crashes again. Thanks to everyone so far who is replying so quickly :-) Cheers, John. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Mar 8 4:18:58 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 835BF37B401 for ; Sat, 8 Mar 2003 04:18:56 -0800 (PST) Received: from murmeldjur.it.su.se (murmeldjur.it.su.se [130.237.95.79]) by mx1.FreeBSD.org (Postfix) with ESMTP id 60FC543F85 for ; Sat, 8 Mar 2003 04:18:55 -0800 (PST) (envelope-from rnyberg@murmeldjur.it.su.se) Received: from murmeldjur.it.su.se (localhost [127.0.0.1]) by murmeldjur.it.su.se (8.12.8/8.12.7) with ESMTP id h28CIrYs059444 for ; Sat, 8 Mar 2003 13:18:53 +0100 (CET) (envelope-from rnyberg@murmeldjur.it.su.se) Received: (from rnyberg@localhost) by murmeldjur.it.su.se (8.12.8/8.12.8/Submit) id h28CIrJl059443 for freebsd-smp@freebsd.org; Sat, 8 Mar 2003 13:18:53 +0100 (CET) Date: Sat, 8 Mar 2003 13:18:52 +0100 From: Richard Nyberg To: freebsd-smp@freebsd.org Subject: hyperthreading randomness Message-ID: <20030308121852.GA25380@murmeldjur.it.su.se> Mail-Followup-To: Richard Nyberg , freebsd-smp@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4i 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 a Dell Precision 450 with 2 xeon CPU:s. The weird thing is that it randomly boots up with either 2 or 4 CPUs on the same 4-STABLE kernel. 2 more often than 4. It seems a bit unpredictable :( CPU: Intel(R) Xeon(TM) CPU 2.66GHz (2657.82-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0xf27 Stepping = 7 Features=0xbfebfbff Hyperthreading: 2 logical CPUs Diff of dmesg with 2 and 4 CPUs. --- /tmp/dmesg.old Sat Mar 8 12:10:33 2003 +++ /var/run/dmesg.boot Sat Mar 8 11:57:38 2003 @@ -19,7 +19,9 @@ Programming 24 pins in IOAPIC #2 FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 0, version: 0x00050014, at 0xfee00000 - cpu1 (AP): apic id: 3, version: 0x00050014, at 0xfee00000 + cpu1 (AP): apic id: 1, version: 0x00050014, at 0xfee00000 + cpu2 (AP): apic id: 2, version: 0x00050014, at 0xfee00000 + cpu3 (AP): apic id: 3, version: 0x00050014, at 0xfee00000 io0 (APIC): apic id: 4, version: 0x00178020, at 0xfec00000 io1 (APIC): apic id: 5, version: 0x00178020, at 0xfec80000 io2 (APIC): apic id: 6, version: 0x00178020, at 0xfec80800 @@ -103,6 +105,8 @@ APIC_IO: Testing 8254 interrupt delivery APIC_IO: routing 8254 via IOAPIC #0 intpin 2 SMP: AP CPU #1 Launched! +SMP: AP CPU #3 Launched! +SMP: AP CPU #2 Launched! ad0: 38146MB [77504/16/63] at ata0-master UDMA100 acd0: DVD-ROM at ata1-master UDMA33 acd1: CD-RW at ata1-slave UDMA33 -Richard To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Mar 8 5:48:31 2003 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id AFA0737B401; Sat, 8 Mar 2003 05:48:29 -0800 (PST) Received: from chunky.global.net.uk (chunky.global.net.uk [80.189.91.76]) by mx1.FreeBSD.org (Postfix) with ESMTP id AC81743FBF; Sat, 8 Mar 2003 05:48:28 -0800 (PST) (envelope-from john.ekins@brightview.com) Received: from ravms by chunky.global.net.uk with mail-ok (Exim 3.36 #1) id 18rec3-0002Dl-00; Sat, 08 Mar 2003 13:43:43 +0000 Received: from [195.137.100.169] (helo=chunky.internal.amaretti.net) by chunky.global.net.uk with smtp (Exim 3.36 #1) id 18rec3-0002De-00; Sat, 08 Mar 2003 13:43:43 +0000 Date: Sat, 8 Mar 2003 13:46:43 +0000 From: John Ekins To: John Baldwin Cc: freebsd-smp@freebsd.org Subject: Re: Regular panic on SMP system Message-Id: <20030308134643.337bf703.john.ekins@brightview.com> In-Reply-To: References: <20030306194400.4fa939e3.john.ekins@brightview.com> Organization: Brightview X-Mailer: Sylpheed version 0.8.6 (GTK+ 1.2.10; i386-portbld-freebsd5.0) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Envelope-To: jhb@FreeBSD.org, freebsd-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 Thu, 06 Mar 2003 15:16:05 -0500 (EST) John Baldwin wrote: -> > Mar 6 10:12:45 mx0 /kernel: Fatal trap 12: page fault while in kernel -> > mode -> > Mar 6 10:12:45 mx0 /kernel: mp_lock = 03000002; cpuid = 3; lapic.id = -> > 03000000 Mar 6 10:12:45 mx0 /kernel: fault virtual address = 0x0 -> -> NULL pointer dereference. Any chance you could get a dump or get a stack -> trace from ddb maybe? OK, the machine has crashed again and I have a dump. Here's the stack trace: (kgdb) where #0 dumpsys () at ../../kern/kern_shutdown.c:487 #1 0xc01749f3 in boot (howto=256) at ../../kern/kern_shutdown.c:316 #2 0xc0174e65 in panic (fmt=0xc0268000 "pmap_release: freeing held page table #page") at ../../kern/kern_shutdown.c:595 #3 0xc02240cc in pmap_release_free_page (pmap=0xe60da82c, p=0xc0ccb724) at ../../i386/i386/pmap.c:1209 #4 0xc02242ce in pmap_release (pmap=0xe60da82c) at #../../i386/i386/pmap.c:1389 5 0xc01e6309 in vmspace_free (vm=0xe60da7c0) at #../../vm/vm_map.c:217 6 0xc022c149 in cpu_wait (p=0xe6a6e820) at #../../i386/i386/vm_machdep.c:271 7 0xc016d357 in wait1 (q=0xe6091440, #uap=0xe60cbf80, compat=0) at ../../kern/kern_exit.c:533 #8 0xc016d0a0 in wait4 (p=0xe6091440, uap=0xe60cbf80) at #../../kern/kern_exit.c:405 9 0xc0227e51 in syscall2 (frame={tf_fs = 47, #tf_es = 47, tf_ds = 47, tf_edi = 134953605, tf_esi = 3600, tf_ebp = -1077940324, tf_isp = -435372076, tf_ebx = 672638540, tf_edx = -1077940356, tf_ecx = 13, tf_eax = 7, tf_trapno = 12, tf_err = 2, tf_eip = 672325272, tf_cs = 31, tf_eflags = 514, tf_esp = -1077940368, tf_ss = 47}) at ../../i386/i386/trap.c:1175 #10 0xc02126ab in Xint0x80_syscall () #11 0x804e5cf in ?? () #12 0x8094744 in ?? () #13 0x8052d67 in ?? () #14 0x80537d5 in ?? () #15 0x80576da in ?? () #16 0x805e85a in ?? () #17 0x804a6f9 in ?? () --- I'm really wondering if this is a hardware thing, because I have several other identical machines with identical kernels and they stay up without problems. Cheers, John. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message