From owner-freebsd-smp@FreeBSD.ORG Mon Jul 14 10:27:17 2003 Return-Path: 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 951CF37B401 for ; Mon, 14 Jul 2003 10:27:17 -0700 (PDT) Received: from delta.ssakhk.cz (delta.ssakhk.cz [195.113.165.66]) by mx1.FreeBSD.org (Postfix) with ESMTP id 192CB43F75 for ; Mon, 14 Jul 2003 10:27:16 -0700 (PDT) (envelope-from veselik@ssakhk.cz) Received: from ssakhk.cz (root@localhost) by delta.ssakhk.cz (8.12.9/8.12.9) with SMTP id h6EHQqN8019205 for ; Mon, 14 Jul 2003 19:26:52 +0200 Received: from VESELIK2 ([192.168.1.187]) by delta.ssakhk.cz (8.12.9/8.12.9) with SMTP id h6EHQpCD019192 for ; Mon, 14 Jul 2003 19:26:51 +0200 From: "Patrik Veselík" To: Date: Mon, 14 Jul 2003 19:27:14 +0200 Message-ID: MIME-Version: 1.0 X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0) X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2727.1300 Importance: Normal X-Antivirus: avast! (VPS 10.7.2003), Outbound message X-Antivirus-Status: Clean Content-Type: text/plain; charset="iso-8859-2" Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.1 Subject: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 14 Jul 2003 17:27:18 -0000 Dear sirs, I create an article about a new chipset supporting more RAM memory than = 4 GB. What is maximum of supported processors of 32-bit x86 family? Or = if its possible - may I get list of computers? Best regards, Patrik Veselik Stredni skola aplikovane kybernetiky, s.r.o. Hradecka 1151 500 02 Hradec Kralove Czech Republic veselik@ssakhk.cz patrik.veselik@cpress.cz veselik@centrum.cz ICQ UIN 13299106 tel. +420 - 495 518 777 fax +420 - 495 518 778 mobil +420 - 608 - 238 328 --- Odchoz=ED zpr=E1va neobsahuje viry. Zkontrolov=E1no antivirov=FDm syst=E9mem AVG (http://www.grisoft.cz). Verze: 6.0.500 / Virov=E1 b=E1ze: 298 - datum vyd=E1n=ED: 10.7.2003 From owner-freebsd-smp@FreeBSD.ORG Mon Jul 14 11:15:07 2003 Return-Path: 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 017C837B446 for ; Mon, 14 Jul 2003 11:15:07 -0700 (PDT) Received: from ztxmail03.ztx.compaq.com (ztxmail03.ztx.compaq.com [161.114.1.207]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2496043F3F for ; Mon, 14 Jul 2003 11:15:06 -0700 (PDT) (envelope-from john.cagle@hp.com) Received: from cceexg12.americas.cpqcorp.net (cceexg12.americas.cpqcorp.net [16.110.250.124]) by ztxmail03.ztx.compaq.com (Postfix) with ESMTP id 990CB9E54; Mon, 14 Jul 2003 13:15:05 -0500 (CDT) Received: from cceexc19.americas.cpqcorp.net ([16.110.250.85]) by cceexg12.americas.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673); Mon, 14 Jul 2003 13:15:01 -0500 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="iso-8859-1" Content-Transfer-Encoding: quoted-printable Date: Mon, 14 Jul 2003 13:15:01 -0500 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: maximum of CPUs Thread-Index: AcNKLTfD/4mTnVBUTJOK1OEEh1oj4QABhaMA From: "Cagle, John (ISS-Houston)" To: =?iso-8859-1?Q?Patrik_Vesel=EDk?= X-OriginalArrivalTime: 14 Jul 2003 18:15:01.0853 (UTC) FILETIME=[D7A9E8D0:01C34A33] cc: freebsd-smp@freebsd.org Subject: RE: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 14 Jul 2003 18:15:07 -0000 I think that 8 CPUs is pretty much the limit for FreeBSD. We have = FreeBSD 5.1 running on an 8-way HP ProLiant DL740 and HP ProLiant DL760. = (That's 8 processors *without* using hyperthreading.) It also runs = very well on an HP ProLiant DL560 which has 4 physical processors (which = is 8 logical CPU's when hyperthreading is turned on). -------------------------------- John Cagle john.cagle@hp.com Principal Member Technical Staff Industry Standard Servers http://www.hp.com/linux/ Hewlett-Packard Company 281-514-9167 > -----Original Message----- > From: Patrik Vesel=EDk [mailto:veselik@ssakhk.cz] > Sent: Monday, July 14, 2003 12:27 PM > To: freebsd-smp@freebsd.org > Subject: maximum of CPUs > > > Dear sirs, > > I create an article about a new chipset supporting more RAM > memory than 4 GB. What is maximum of supported processors of > 32-bit x86 family? Or if its possible - may I get list of computers? > > Best regards, > > Patrik Veselik > Stredni skola aplikovane kybernetiky, s.r.o. > Hradecka 1151 > 500 02 Hradec Kralove > Czech Republic > veselik@ssakhk.cz > patrik.veselik@cpress.cz > veselik@centrum.cz > ICQ UIN 13299106 > tel. +420 - 495 518 777 > fax +420 - 495 518 778 > mobil +420 - 608 - 238 328 > > --- > Odchoz=ED zpr=E1va neobsahuje viry. > Zkontrolov=E1no antivirov=FDm syst=E9mem AVG (http://www.grisoft.cz). > Verze: 6.0.500 / Virov=E1 b=E1ze: 298 - datum vyd=E1n=ED: 10.7.2003 > _______________________________________________ > freebsd-smp@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-smp > To unsubscribe, send any mail to "freebsd-smp-unsubscribe@freebsd.org" >=20 From owner-freebsd-smp@FreeBSD.ORG Mon Jul 14 12:58:51 2003 Return-Path: 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 4D27737B401 for ; Mon, 14 Jul 2003 12:58:51 -0700 (PDT) Received: from scribble.fsn.hu (scribble.fsn.hu [193.224.40.95]) by mx1.FreeBSD.org (Postfix) with SMTP id A1E4243F85 for ; Mon, 14 Jul 2003 12:58:49 -0700 (PDT) (envelope-from bra@fsn.hu) Received: (qmail 22863 invoked from network); 14 Jul 2003 19:58:46 -0000 Received: from unknown (HELO fsn.hu) (127.0.0.1) by localhost with SMTP; 14 Jul 2003 19:58:46 -0000 Message-ID: <3F130B72.5000701@fsn.hu> Date: Mon, 14 Jul 2003 21:58:42 +0200 From: Attila Nagy Organization: Free Software Network Foundation - http://www.fsn.hu/ User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030529 X-Accept-Language: hu, en-us, en MIME-Version: 1.0 To: "Cagle, John (ISS-Houston)" References: In-Reply-To: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: freebsd-smp@freebsd.org cc: =?ISO-8859-1?Q?Patrik_Vesel=EDk?= Subject: Re: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 14 Jul 2003 19:58:51 -0000 Cagle, John (ISS-Houston) wrote: > I think that 8 CPUs is pretty much the limit for FreeBSD. > We have FreeBSD 5.1 running on an 8-way HP ProLiant DL740 and HP > ProLiant DL760. (That's 8 processors *without* using hyperthreading.) Just out of curiosity: what happens if you enable hyperthreading? -- Attila Nagy e-mail: Attila.Nagy@fsn.hu Free Software Network (FSN.HU) phone @work: +361 210 1415/127 ISOs: http://www.fsn.hu/?f=download cell.: +3630 306 6758 From owner-freebsd-smp@FreeBSD.ORG Mon Jul 14 13:06:30 2003 Return-Path: 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 AF36037B401 for ; Mon, 14 Jul 2003 13:06:29 -0700 (PDT) Received: from ztxmail03.ztx.compaq.com (ztxmail03.ztx.compaq.com [161.114.1.207]) by mx1.FreeBSD.org (Postfix) with ESMTP id E855943F75 for ; Mon, 14 Jul 2003 13:06:28 -0700 (PDT) (envelope-from john.cagle@hp.com) Received: from cceexg11.americas.cpqcorp.net (cceexg11.americas.cpqcorp.net [16.110.250.125]) by ztxmail03.ztx.compaq.com (Postfix) with ESMTP id 82419A2A3; Mon, 14 Jul 2003 15:06:28 -0500 (CDT) Received: from cceexc19.americas.cpqcorp.net ([16.110.250.85]) by cceexg11.americas.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673); Mon, 14 Jul 2003 15:06:28 -0500 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="iso-8859-1" Content-Transfer-Encoding: quoted-printable Date: Mon, 14 Jul 2003 15:05:48 -0500 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: maximum of CPUs Thread-Index: AcNKQqDRGkdZ1euPRnS2DgHd/+8aXQAAFqJg From: "Cagle, John (ISS-Houston)" To: "Attila Nagy" X-OriginalArrivalTime: 14 Jul 2003 20:06:28.0398 (UTC) FILETIME=[6927B4E0:01C34A43] cc: freebsd-smp@freebsd.org cc: =?iso-8859-1?Q?Patrik_Vesel=EDk?= Subject: RE: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 14 Jul 2003 20:06:30 -0000 If you enable hyperthreading on an 8-way server, you'll end up with 16 = logical processors (and a *whole bunch* of APIC's) -- a scenario that is = not comprehended by FreeBSD. In other words, there needs to be some = more engineering done on FreeBSD to support more than 8 processors = (AFAICT). > -----Original Message----- > From: Attila Nagy [mailto:bra@fsn.hu]=20 > Sent: Monday, July 14, 2003 2:59 PM > To: Cagle, John (ISS-Houston) > Cc: Patrik Vesel=EDk; freebsd-smp@freebsd.org > Subject: Re: maximum of CPUs >=20 >=20 > Cagle, John (ISS-Houston) wrote: > > I think that 8 CPUs is pretty much the limit for FreeBSD. > > We have FreeBSD 5.1 running on an 8-way HP ProLiant DL740 and HP > > ProLiant DL760. (That's 8 processors *without* using=20 > hyperthreading.) > Just out of curiosity: what happens if you enable hyperthreading? From owner-freebsd-smp@FreeBSD.ORG Mon Jul 14 13:53:10 2003 Return-Path: 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 ABADA37B401 for ; Mon, 14 Jul 2003 13:53:10 -0700 (PDT) Received: from scribble.fsn.hu (scribble.fsn.hu [193.224.40.95]) by mx1.FreeBSD.org (Postfix) with SMTP id 59EE643FA3 for ; Mon, 14 Jul 2003 13:53:09 -0700 (PDT) (envelope-from bra@fsn.hu) Received: (qmail 24324 invoked from network); 14 Jul 2003 20:53:07 -0000 Received: from unknown (HELO fsn.hu) (127.0.0.1) by localhost with SMTP; 14 Jul 2003 20:53:07 -0000 Message-ID: <3F13182B.9040209@fsn.hu> Date: Mon, 14 Jul 2003 22:52:59 +0200 From: Attila Nagy Organization: Free Software Network Foundation - http://www.fsn.hu/ User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030529 X-Accept-Language: hu, en-us, en MIME-Version: 1.0 To: "Cagle, John (ISS-Houston)" References: In-Reply-To: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: freebsd-smp@freebsd.org cc: =?ISO-8859-1?Q?Patrik_Vesel=EDk?= Subject: Re: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 14 Jul 2003 20:53:11 -0000 Cagle, John (ISS-Houston) wrote: > If you enable hyperthreading on an 8-way server, you'll end up with 16 logical processors (and a *whole bunch* of APIC's) -- a scenario that is not comprehended by FreeBSD. In other words, there needs to be some more engineering done on FreeBSD to support more than 8 processors (AFAICT). I remember now about the discussions about this. Thanks. -- Attila Nagy e-mail: Attila.Nagy@fsn.hu Free Software Network (FSN.HU) phone @work: +361 210 1415/127 ISOs: http://www.fsn.hu/?f=download cell.: +3630 306 6758 From owner-freebsd-smp@FreeBSD.ORG Mon Jul 14 14:49:57 2003 Return-Path: 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 7740737B404 for ; Mon, 14 Jul 2003 14:49:57 -0700 (PDT) Received: from mail.speakeasy.net (mail12.speakeasy.net [216.254.0.212]) by mx1.FreeBSD.org (Postfix) with ESMTP id 85B8543FAF for ; Mon, 14 Jul 2003 14:49:56 -0700 (PDT) (envelope-from jhb@FreeBSD.org) Received: (qmail 29533 invoked from network); 14 Jul 2003 21:49:55 -0000 Received: from unknown (HELO server.baldwin.cx) ([216.27.160.63]) (envelope-sender )encrypted SMTP for ; 14 Jul 2003 21:49:55 -0000 Received: from laptop.baldwin.cx (gw1.twc.weather.com [216.133.140.1]) by server.baldwin.cx (8.12.9/8.12.9) with ESMTP id h6ELnoGI017962; Mon, 14 Jul 2003 17:49:51 -0400 (EDT) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.5.4 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, 14 Jul 2003 17:50:06 -0400 (EDT) From: John Baldwin To: "Cagle, John (ISS-Houston)" cc: freebsd-smp@freebsd.org cc: =?iso-8859-1?Q?Patrik_Vesel=EDk?= Subject: RE: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 14 Jul 2003 21:49:57 -0000 On 14-Jul-2003 Cagle, John (ISS-Houston) wrote: > If you enable hyperthreading on an 8-way server, you'll end up with 16 logical processors (and a > *whole bunch* of APIC's) -- a scenario that is not comprehended by FreeBSD. In other words, > there needs to be some more engineering done on FreeBSD to support more than 8 processors > (AFAICT). Well, the problem is that the original APIC's only supported 15 physical APIC IDs. The (x)APIC's from Pentium4+ support 255 physical APIC IDs. If the BIOS does the right thing as far as providing ID's for the I/O APICs in the mptable and/or ACPI MADT, future versions of FreeBSD should work just fine. -- John Baldwin <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ From owner-freebsd-smp@FreeBSD.ORG Tue Jul 15 07:09:48 2003 Return-Path: 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 DD82C37B401; Tue, 15 Jul 2003 07:09:47 -0700 (PDT) Received: from ztxmail04.ztx.compaq.com (ztxmail04.ztx.compaq.com [161.114.1.208]) by mx1.FreeBSD.org (Postfix) with ESMTP id DCEE943FBF; Tue, 15 Jul 2003 07:09:46 -0700 (PDT) (envelope-from john.cagle@hp.com) Received: from cceexg11.americas.cpqcorp.net (cceexg11.americas.cpqcorp.net [16.110.250.125]) by ztxmail04.ztx.compaq.com (Postfix) with ESMTP id 6EC0EBC43; Tue, 15 Jul 2003 09:09:46 -0500 (CDT) Received: from cceexc19.americas.cpqcorp.net ([16.110.250.85]) by cceexg11.americas.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673); Tue, 15 Jul 2003 09:09:45 -0500 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 Date: Tue, 15 Jul 2003 09:09:44 -0500 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: maximum of CPUs Thread-Index: AcNKUe86aWvnFoLzT2yHU0Zljs7bcAAhg08w From: "Cagle, John (ISS-Houston)" To: "John Baldwin" X-OriginalArrivalTime: 15 Jul 2003 14:09:45.0343 (UTC) FILETIME=[BE5A70F0:01C34ADA] cc: freebsd-smp@freebsd.org cc: =?us-ascii?Q?Patrik_Veselik?= Subject: RE: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 15 Jul 2003 14:09:48 -0000 > Well, the problem is that the original APIC's only supported=20 > 15 physical > APIC IDs. The (x)APIC's from Pentium4+ support 255 physical APIC IDs. > If the BIOS does the right thing as far as providing ID's for the I/O > APICs in the mptable and/or ACPI MADT, future versions of=20 > FreeBSD should > work just fine. >=20 > --=20 >=20 > John Baldwin <>< http://www.FreeBSD.org/~jhb/ > "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/=20 John, The BIOS on our 8-way does list all 3 IOAPICs in the mptable (not the local APICs), but it does not list the logical processors, only the physical processors: Processors: APIC ID Version State Family Model Step Flags 0 0x14 BSP, usable 15 2 2 0x3febfbff 2 0x14 AP, usable 15 2 2 0x3febfbff 4 0x14 AP, usable 15 2 2 0x3febfbff 6 0x14 AP, usable 15 2 2 0x3febfbff 8 0x14 AP, usable 15 2 2 0x3febfbff 10 0x14 AP, usable 15 2 2 0x3febfbff 12 0x14 AP, usable 15 2 2 0x3febfbff 14 0x14 AP, usable 15 2 2 0x3febfbff -- I/O APICs: APIC ID Version State Address 8 0x11 usable 0xfec00000 9 0x11 usable 0xfec01000 10 0x11 usable 0xfec02000 Is this what you would expect to see on an 8-way server that supports hyperthreading? Thanks, John P.S. Here is the panic that I get currently: CPU: Intel(R) XEON(TM) MP CPU 1.50GHz (1499.46-MHz 686-class CPU) Origin =3D "GenuineIntel" Id =3D 0xf22 Stepping =3D 2 =20 Features=3D0x3febfbff Hyperthreading: 2 logical CPUs real memory =3D 2147418112 (2047 MB) avail memory =3D 2082279424 (1985 MB) panic: No free physical APIC IDs found cpuid =3D 0; lapic.id =3D 00000000 Debugger("panic") Stopped at Debugger+0x55: xchgl %ebx,in_Debugger.0 db> trace Debugger(c050d3ac,0,c0527df1,c0792d24,100) at Debugger+0x55 panic(c0527df1,8,1,5,c3b083cc) at panic+0x13f fix_mp_table(0,c059cb50,c0792d80,c0499b1e,c0527846) at fix_mp_table+0x1b5 mp_enable(9f000,c0792d80,c0334901,c060a060,c0510056) at mp_enable+0x5a cpu_mp_start(c060a060,c0510056,0,1,c0792d98) at cpu_mp_start+0x3d mp_start(0,78f000,78fc00,78f000,0) at mp_start+0x41 mi_startup() at mi_startup+0xb5 begin() at begin+0x2c From owner-freebsd-smp@FreeBSD.ORG Tue Jul 15 10:14:32 2003 Return-Path: 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 AB2F837B401 for ; Tue, 15 Jul 2003 10:14:32 -0700 (PDT) Received: from pirx.hexapodia.org (pirx.hexapodia.org [208.42.114.113]) by mx1.FreeBSD.org (Postfix) with ESMTP id BBA2543F75 for ; Tue, 15 Jul 2003 10:14:31 -0700 (PDT) (envelope-from adi@hexapodia.org) Received: by pirx.hexapodia.org (Postfix, from userid 22448) id 25AFCB422; Tue, 15 Jul 2003 12:14:30 -0500 (CDT) Date: Tue, 15 Jul 2003 12:14:29 -0500 From: Andy Isaacson To: =?iso-8859-1?Q?Patrik_Vesel=EDk?= Message-ID: <20030715121429.A11267@hexapodia.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit User-Agent: Mutt/1.2.5i In-Reply-To: ; from veselik@ssakhk.cz on Mon, Jul 14, 2003 at 07:27:14PM +0200 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 cc: freebsd-smp@freebsd.org Subject: Re: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 15 Jul 2003 17:14:32 -0000 On Mon, Jul 14, 2003 at 07:27:14PM +0200, Patrik Veselík wrote: > I create an article about a new chipset supporting more RAM memory > than 4 GB. What is maximum of supported processors of 32-bit x86 > family? Or if its possible - may I get list of computers? I can't tell if you're talking about "supported by FreeBSD" or "supported running any OS". Several vendors sell boxes with 32 x86 processors, including IBM (the part that used to be Sequent), Unisys, and one of the Japanese vendors (NEC or Fujitsu or something like that). The easiest to find links to is http://www.unisys.com/products/es7000__servers/hardware/index.htm -andy From owner-freebsd-smp@FreeBSD.ORG Tue Jul 15 12:09:32 2003 Return-Path: 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 7E48537B401 for ; Tue, 15 Jul 2003 12:09:32 -0700 (PDT) Received: from scribble.fsn.hu (scribble.fsn.hu [193.224.40.95]) by mx1.FreeBSD.org (Postfix) with SMTP id 0CE8D43F3F for ; Tue, 15 Jul 2003 12:09:31 -0700 (PDT) (envelope-from bra@fsn.hu) Received: (qmail 15431 invoked from network); 15 Jul 2003 19:09:27 -0000 Received: from unknown (HELO fsn.hu) (127.0.0.1) by localhost with SMTP; 15 Jul 2003 19:09:27 -0000 Message-ID: <3F145157.4050009@fsn.hu> Date: Tue, 15 Jul 2003 21:09:11 +0200 From: Attila Nagy Organization: Free Software Network Foundation - http://www.fsn.hu/ User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030529 X-Accept-Language: hu, en-us, en MIME-Version: 1.0 To: Andy Isaacson References: <20030715121429.A11267@hexapodia.org> In-Reply-To: <20030715121429.A11267@hexapodia.org> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: freebsd-smp@freebsd.org cc: =?ISO-8859-1?Q?Patrik_Vesel=EDk?= Subject: Re: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 15 Jul 2003 19:09:32 -0000 Andy Isaacson wrote: > I can't tell if you're talking about "supported by FreeBSD" or > "supported running any OS". Several vendors sell boxes with 32 x86 > processors, including IBM (the part that used to be Sequent), Unisys, > and one of the Japanese vendors (NEC or Fujitsu or something like that). > The easiest to find links to is > http://www.unisys.com/products/es7000__servers/hardware/index.htm Even Linux needs a patch for these machines :) http://lists.insecure.org/lists/linux-kernel/2003/Jun/2612.html AFAIK until this patch, only Windows 2000 was officially supported by Unisys. -- Attila Nagy e-mail: Attila.Nagy@fsn.hu Free Software Network (FSN.HU) phone @work: +361 210 1415/127 ISOs: http://www.fsn.hu/?f=download cell.: +3630 306 6758 From owner-freebsd-smp@FreeBSD.ORG Tue Jul 15 12:34:38 2003 Return-Path: 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 5396037B401 for ; Tue, 15 Jul 2003 12:34:38 -0700 (PDT) Received: from mail.speakeasy.net (mail14.speakeasy.net [216.254.0.214]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9668C43F93 for ; Tue, 15 Jul 2003 12:34:37 -0700 (PDT) (envelope-from jhb@FreeBSD.org) Received: (qmail 15612 invoked from network); 15 Jul 2003 19:34:36 -0000 Received: from unknown (HELO server.baldwin.cx) ([216.27.160.63]) (envelope-sender )encrypted SMTP for ; 15 Jul 2003 19:34:36 -0000 Received: from laptop.baldwin.cx (gw1.twc.weather.com [216.133.140.1]) by server.baldwin.cx (8.12.9/8.12.9) with ESMTP id h6FJYYGI020684; Tue, 15 Jul 2003 15:34:35 -0400 (EDT) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.5.4 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: Tue, 15 Jul 2003 15:34:50 -0400 (EDT) From: John Baldwin To: "Cagle, John (ISS-Houston)" cc: freebsd-smp@freebsd.org cc: =?us-ascii?Q?Patrik_Veselik?= Subject: RE: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 15 Jul 2003 19:34:38 -0000 On 15-Jul-2003 Cagle, John (ISS-Houston) wrote: >> Well, the problem is that the original APIC's only supported >> 15 physical >> APIC IDs. The (x)APIC's from Pentium4+ support 255 physical APIC IDs. >> If the BIOS does the right thing as far as providing ID's for the I/O >> APICs in the mptable and/or ACPI MADT, future versions of >> FreeBSD should >> work just fine. >> >> -- >> >> John Baldwin <>< http://www.FreeBSD.org/~jhb/ >> "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ > > John, > > The BIOS on our 8-way does list all 3 IOAPICs in the mptable (not the > local APICs), but it does not list the logical processors, only the > physical processors: > > Processors: APIC ID Version State Family Model Step > Flags > 0 0x14 BSP, usable 15 2 2 > 0x3febfbff > 2 0x14 AP, usable 15 2 2 > 0x3febfbff > 4 0x14 AP, usable 15 2 2 > 0x3febfbff > 6 0x14 AP, usable 15 2 2 > 0x3febfbff > 8 0x14 AP, usable 15 2 2 > 0x3febfbff > 10 0x14 AP, usable 15 2 2 > 0x3febfbff > 12 0x14 AP, usable 15 2 2 > 0x3febfbff > 14 0x14 AP, usable 15 2 2 > 0x3febfbff > -- > I/O APICs: APIC ID Version State Address > 8 0x11 usable 0xfec00000 > 9 0x11 usable 0xfec01000 > 10 0x11 usable 0xfec02000 > > Is this what you would expect to see on an 8-way server that supports > hyperthreading? Ugh, I would hope that the APIC ID's wouldn't collide with CPUs. :( Note that you have real CPU's at ID's 8 and 10 and a logical one at 9. Currently my code doesn't renumber APIC ID's to try to cope with this type of case. Does the ACPI MADT table report the same values for the APIC ID's of the I/O APICs? > P.S. Here is the panic that I get currently: > > CPU: Intel(R) XEON(TM) MP CPU 1.50GHz (1499.46-MHz 686-class CPU) > Origin = "GenuineIntel" Id = 0xf22 Stepping = 2 > > Features=0x3febfbff ,MCA,C > MOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM> > Hyperthreading: 2 logical CPUs > real memory = 2147418112 (2047 MB) > avail memory = 2082279424 (1985 MB) > panic: No free physical APIC IDs found This is from the current code that is trying to renumber the I/O APIC's since they collide with CPUs. I've completely rewritten all the mptable code, but I don't support renumbering I/O APICs yet. I was really hoping that the BIOS would help us out enough to keep the IDs from overlapping. -- John Baldwin <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ From owner-freebsd-smp@FreeBSD.ORG Tue Jul 15 23:04:15 2003 Return-Path: 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 6D69E37B401; Tue, 15 Jul 2003 23:04:15 -0700 (PDT) Received: from stork.mail.pas.earthlink.net (stork.mail.pas.earthlink.net [207.217.120.188]) by mx1.FreeBSD.org (Postfix) with ESMTP id BFFD543FA3; Tue, 15 Jul 2003 23:04:14 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from user-38lc1c9.dialup.mindspring.com ([209.86.5.137] helo=mindspring.com) by stork.mail.pas.earthlink.net with asmtp (SSLv3:RC4-MD5:128) (Exim 3.33 #1) id 19cfOf-0000lm-00; Tue, 15 Jul 2003 23:04:13 -0700 Message-ID: <3F14EA9E.616847E1@mindspring.com> Date: Tue, 15 Jul 2003 23:03:10 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: John Baldwin References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-ELNK-Trace: b1a02af9316fbb217a47c185c03b154d40683398e744b8a45ad0c910816b1f9de3d5133016723803350badd9bab72f9c350badd9bab72f9c350badd9bab72f9c cc: freebsd-smp@freebsd.org cc: Patrik Veselik Subject: Re: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Jul 2003 06:04:15 -0000 > This is from the current code that is trying to renumber the I/O > APIC's since they collide with CPUs. I've completely rewritten > all the mptable code, but I don't support renumbering I/O APICs > yet. I was really hoping that the BIOS would help us out enough > to keep the IDs from overlapping. I thought it had to, to be MP Spec 1.4 compliant?!? This whole "stuff everything into ACPI and to hell with the standards" thing that vendors are doing on Intel these days is really, really annoying. -- Terry From owner-freebsd-smp@FreeBSD.ORG Tue Jul 15 23:36:05 2003 Return-Path: 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 F108137B401 for ; Tue, 15 Jul 2003 23:36:04 -0700 (PDT) Received: from stork.mail.pas.earthlink.net (stork.mail.pas.earthlink.net [207.217.120.188]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5C8AE43F3F for ; Tue, 15 Jul 2003 23:36:04 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from user-38lc1c9.dialup.mindspring.com ([209.86.5.137] helo=mindspring.com) by stork.mail.pas.earthlink.net with asmtp (SSLv3:RC4-MD5:128) (Exim 3.33 #1) id 19cftL-0004WG-00; Tue, 15 Jul 2003 23:35:56 -0700 Message-ID: <3F14F20D.605BEBD5@mindspring.com> Date: Tue, 15 Jul 2003 23:34:53 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Attila Nagy References: <20030715121429.A11267@hexapodia.org> <3F145157.4050009@fsn.hu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-ELNK-Trace: b1a02af9316fbb217a47c185c03b154d40683398e744b8a47b9cbb4dea58817fefa48ab48c9021dda8438e0f32a48e08350badd9bab72f9c350badd9bab72f9c cc: freebsd-smp@freebsd.org cc: Patrik =?iso-8859-1?Q?Vesel=EDk?= Subject: Re: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Jul 2003 06:36:05 -0000 Attila Nagy wrote: > Andy Isaacson wrote: > > I can't tell if you're talking about "supported by FreeBSD" or > > "supported running any OS". Several vendors sell boxes with 32 x86 > > processors, including IBM (the part that used to be Sequent), Unisys, > > and one of the Japanese vendors (NEC or Fujitsu or something like that). > > The easiest to find links to is > > http://www.unisys.com/products/es7000__servers/hardware/index.htm > > Even Linux needs a patch for these machines :) > http://lists.insecure.org/lists/linux-kernel/2003/Jun/2612.html > > AFAIK until this patch, only Windows 2000 was officially supported by > Unisys. The Sequent Symmetry (i386 based) was the same machine as the Unisys 5000/20, which was an OEM version of the box. I was pretty sure that the ES7000/40 was also an OEM Sequent box? If so, it may be that if they didn't put too much vendor BIOS voodoo in the things, this would also let them run on matching Sequent badged hardware... and that's some pretty cool iron, even these days. -- Terry From owner-freebsd-smp@FreeBSD.ORG Wed Jul 16 00:12:47 2003 Return-Path: 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 83EF937B401; Wed, 16 Jul 2003 00:12:47 -0700 (PDT) Received: from ztxmail04.ztx.compaq.com (ztxmail04.ztx.compaq.com [161.114.1.208]) by mx1.FreeBSD.org (Postfix) with ESMTP id BD9A543F85; Wed, 16 Jul 2003 00:12:46 -0700 (PDT) (envelope-from john.cagle@hp.com) Received: from cceexg13.americas.cpqcorp.net (cceexg13.americas.cpqcorp.net [16.110.250.119]) by ztxmail04.ztx.compaq.com (Postfix) with ESMTP id 4891FBED3; Wed, 16 Jul 2003 02:12:46 -0500 (CDT) Received: from cceexc19.americas.cpqcorp.net ([16.110.250.85]) by cceexg13.americas.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673); Wed, 16 Jul 2003 02:12:46 -0500 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 Date: Wed, 16 Jul 2003 02:12:45 -0500 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: maximum of CPUs Thread-Index: AcNLCFTjQFJs6gFBQaOUmgGcKBOQGQAYIUeA From: "Cagle, John (ISS-Houston)" To: "John Baldwin" X-OriginalArrivalTime: 16 Jul 2003 07:12:46.0170 (UTC) FILETIME=[A82D5BA0:01C34B69] cc: freebsd-smp@freebsd.org cc: Patrik Veselik Subject: RE: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Jul 2003 07:12:47 -0000 > -----Original Message----- > From: John Baldwin [mailto:jhb@FreeBSD.org]=20 > On 15-Jul-2003 Cagle, John (ISS-Houston) wrote: > > -- > > I/O APICs: APIC ID Version State Address > > 8 0x11 usable 0xfec00000 > > 9 0x11 usable 0xfec01000 > > 10 0x11 usable 0xfec02000 > >=20 > > Is this what you would expect to see on an 8-way server=20 > that supports > > hyperthreading? >=20 > Ugh, I would hope that the APIC ID's wouldn't collide with CPUs. :( > Note that you have real CPU's at ID's 8 and 10 and a logical one at > 9. Currently my code doesn't renumber APIC ID's to try to cope with > this type of case. Does the ACPI MADT table report the same values > for the APIC ID's of the I/O APICs? I should have updated my BIOS before trying this out. Sure enough, this was a BIOS bug that was corrected in March. The I/O APICS are now numbered 16, 17 & 18 so as to not collide with processor's local APICS. (Looks like the original BIOS only accounted for processors 0..7, without hyperthreading.) However, even with the properly renumbered IOAPICs, I still get the same panic: "No free physical APIC IDs found". I'm going to add some debug messages and track this down further. I'll let you know if I find anything. I'll also check the MADT to see if it jives with everything else. Thanks, John From owner-freebsd-smp@FreeBSD.ORG Wed Jul 16 08:56:31 2003 Return-Path: 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 25CD037B401 for ; Wed, 16 Jul 2003 08:56:31 -0700 (PDT) Received: from pirx.hexapodia.org (pirx.hexapodia.org [208.42.114.113]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1875743F3F for ; Wed, 16 Jul 2003 08:56:30 -0700 (PDT) (envelope-from adi@hexapodia.org) Received: by pirx.hexapodia.org (Postfix, from userid 22448) id 436F4B433; Wed, 16 Jul 2003 10:56:29 -0500 (CDT) Date: Wed, 16 Jul 2003 10:56:29 -0500 From: Andy Isaacson To: Terry Lambert Message-ID: <20030716105629.A21686@hexapodia.org> References: <20030715121429.A11267@hexapodia.org> <3F145157.4050009@fsn.hu> <3F14F20D.605BEBD5@mindspring.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <3F14F20D.605BEBD5@mindspring.com>; from tlambert2@mindspring.com on Tue, Jul 15, 2003 at 11:34:53PM -0700 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 cc: freebsd-smp@freebsd.org cc: =?iso-8859-1?Q?Patrik_Vesel=EDk?= Subject: Re: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Jul 2003 15:56:31 -0000 On Tue, Jul 15, 2003 at 11:34:53PM -0700, Terry Lambert wrote: > Attila Nagy wrote: > > Andy Isaacson wrote: > > > I can't tell if you're talking about "supported by FreeBSD" or > > > "supported running any OS". Several vendors sell boxes with 32 x86 > > > processors, including IBM (the part that used to be Sequent), Unisys, > > > and one of the Japanese vendors (NEC or Fujitsu or something like that). > > > The easiest to find links to is > > > http://www.unisys.com/products/es7000__servers/hardware/index.htm > > > > Even Linux needs a patch for these machines :) > > http://lists.insecure.org/lists/linux-kernel/2003/Jun/2612.html > > > > AFAIK until this patch, only Windows 2000 was officially supported by > > Unisys. > > The Sequent Symmetry (i386 based) was the same machine as the > Unisys 5000/20, which was an OEM version of the box. I was > pretty sure that the ES7000/40 was also an OEM Sequent box? > > If so, it may be that if they didn't put too much vendor BIOS > voodoo in the things, this would also let them run on matching > Sequent badged hardware... and that's some pretty cool iron, > even these days. Hmm, I thought the 7000 series is Unisys-designed, but further research makes me uncertain. IBM nee Sequent has a serious effort going to get Linux running well on their 8- to 32-CPU boxes. The Unisys effort seems much more lackadaisical. IBM is looking to run single large database images (and whatnot) on theirs, while Unisys seems to be content doing server consolidation. So IBM is working on large SMP scaling issues, while Sequent is just adding the necessary BIOS/APIC/ppb support. Note that Unisys also supports Linux on their mainframe-hybrid boxes like the ClearPath: http://www.unisys.com/products/clearpath__servers/clearpath__plus__os__2200/models.htm -andy From owner-freebsd-smp@FreeBSD.ORG Wed Jul 16 11:59:59 2003 Return-Path: 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 9AB0437B401 for ; Wed, 16 Jul 2003 11:59:59 -0700 (PDT) Received: from mail.speakeasy.net (mail16.speakeasy.net [216.254.0.216]) by mx1.FreeBSD.org (Postfix) with ESMTP id E79EF43FAF for ; Wed, 16 Jul 2003 11:59:58 -0700 (PDT) (envelope-from jhb@FreeBSD.org) Received: (qmail 16792 invoked from network); 16 Jul 2003 18:59:58 -0000 Received: from unknown (HELO server.baldwin.cx) ([216.27.160.63]) (envelope-sender )encrypted SMTP for ; 16 Jul 2003 18:59:58 -0000 Received: from laptop.baldwin.cx (gw1.twc.weather.com [216.133.140.1]) by server.baldwin.cx (8.12.9/8.12.9) with ESMTP id h6GIxuGI023507; Wed, 16 Jul 2003 14:59:56 -0400 (EDT) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.5.4 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: Wed, 16 Jul 2003 15:00:12 -0400 (EDT) From: John Baldwin To: "Cagle, John (ISS-Houston)" cc: freebsd-smp@freebsd.org cc: Patrik Veselik Subject: RE: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Jul 2003 18:59:59 -0000 On 16-Jul-2003 Cagle, John (ISS-Houston) wrote: >> -----Original Message----- >> From: John Baldwin [mailto:jhb@FreeBSD.org] >> On 15-Jul-2003 Cagle, John (ISS-Houston) wrote: > >> > -- >> > I/O APICs: APIC ID Version State Address >> > 8 0x11 usable 0xfec00000 >> > 9 0x11 usable 0xfec01000 >> > 10 0x11 usable 0xfec02000 >> > >> > Is this what you would expect to see on an 8-way server >> that supports >> > hyperthreading? >> >> Ugh, I would hope that the APIC ID's wouldn't collide with CPUs. :( >> Note that you have real CPU's at ID's 8 and 10 and a logical one at >> 9. Currently my code doesn't renumber APIC ID's to try to cope with >> this type of case. Does the ACPI MADT table report the same values >> for the APIC ID's of the I/O APICs? > > I should have updated my BIOS before trying this out. Sure enough, this > was a BIOS bug that was corrected in March. The I/O APICS are now > numbered 16, 17 & 18 so as to not collide with processor's local APICS. > (Looks like the original BIOS only accounted for processors 0..7, > without hyperthreading.) Ok, good. > However, even with the properly renumbered IOAPICs, I still get the same > panic: "No free physical APIC IDs found". This is still expected with the current code. > I'm going to add some debug messages and track this down further. I'll > let you know if I find anything. I'll also check the MADT to see if it > jives with everything else. Tor's suggestion to bump up NAPICID to 256 might help. -- John Baldwin <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ From owner-freebsd-smp@FreeBSD.ORG Wed Jul 16 12:00:04 2003 Return-Path: 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 A9A4D37B401 for ; Wed, 16 Jul 2003 12:00:04 -0700 (PDT) Received: from mail.speakeasy.net (mail8.speakeasy.net [216.254.0.208]) by mx1.FreeBSD.org (Postfix) with ESMTP id 01A3343FAF for ; Wed, 16 Jul 2003 12:00:02 -0700 (PDT) (envelope-from jhb@FreeBSD.org) Received: (qmail 19082 invoked from network); 16 Jul 2003 19:00:01 -0000 Received: from unknown (HELO server.baldwin.cx) ([216.27.160.63]) (envelope-sender )encrypted SMTP for ; 16 Jul 2003 19:00:01 -0000 Received: from laptop.baldwin.cx (gw1.twc.weather.com [216.133.140.1]) by server.baldwin.cx (8.12.9/8.12.9) with ESMTP id h6GIxxGI023510; Wed, 16 Jul 2003 14:59:59 -0400 (EDT) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.5.4 on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <3F14EA9E.616847E1@mindspring.com> Date: Wed, 16 Jul 2003 15:00:15 -0400 (EDT) From: John Baldwin To: Terry Lambert cc: freebsd-smp@freebsd.org cc: Patrik Veselik Subject: Re: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Jul 2003 19:00:05 -0000 On 16-Jul-2003 Terry Lambert wrote: >> This is from the current code that is trying to renumber the I/O >> APIC's since they collide with CPUs. I've completely rewritten >> all the mptable code, but I don't support renumbering I/O APICs >> yet. I was really hoping that the BIOS would help us out enough >> to keep the IDs from overlapping. > > I thought it had to, to be MP Spec 1.4 compliant?!? What had to be what to be compliant? > This whole "stuff everything into ACPI and to hell with the > standards" thing that vendors are doing on Intel these days > is really, really annoying. No, it's correct. The mptable is quite limited and ACPI's configuration mechanism is much cleaner. Complaining about ACPI in this matter only serves to display your ignorance of what it is and how it works. Please educate yourself on the real issues first, thanks. -- John Baldwin <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ From owner-freebsd-smp@FreeBSD.ORG Wed Jul 16 22:59:13 2003 Return-Path: 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 E112337B401; Wed, 16 Jul 2003 22:59:13 -0700 (PDT) Received: from puffin.mail.pas.earthlink.net (puffin.mail.pas.earthlink.net [207.217.120.139]) by mx1.FreeBSD.org (Postfix) with ESMTP id 14F0F43F93; Wed, 16 Jul 2003 22:59:13 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from user-uinj8ql.dialup.mindspring.com ([165.121.163.85] helo=mindspring.com) by puffin.mail.pas.earthlink.net with asmtp (SSLv3:RC4-MD5:128) (Exim 3.33 #1) id 19d1nL-0002Ks-00; Wed, 16 Jul 2003 22:59:12 -0700 Message-ID: <3F163AF2.967C117D@mindspring.com> Date: Wed, 16 Jul 2003 22:58:10 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: John Baldwin References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-ELNK-Trace: b1a02af9316fbb217a47c185c03b154d40683398e744b8a4cb71d4917caf698f9cdc4eacbb235a7a350badd9bab72f9c350badd9bab72f9c350badd9bab72f9c cc: freebsd-smp@freebsd.org cc: Patrik Veselik Subject: Re: maximum of CPUs X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 17 Jul 2003 05:59:14 -0000 John Baldwin wrote: > On 16-Jul-2003 Terry Lambert wrote: > >> This is from the current code that is trying to renumber the I/O > >> APIC's since they collide with CPUs. I've completely rewritten > >> all the mptable code, but I don't support renumbering I/O APICs > >> yet. I was really hoping that the BIOS would help us out enough > >> to keep the IDs from overlapping. > > > > I thought it had to, to be MP Spec 1.4 compliant?!? > > What had to be what to be compliant? I thought the BIOS had to "help us out enough" for the motherboard to be MP Spec 1.4 compliant. > > This whole "stuff everything into ACPI and to hell with the > > standards" thing that vendors are doing on Intel these days > > is really, really annoying. > > No, it's correct. The mptable is quite limited and ACPI's > configuration mechanism is much cleaner. Complaining about > ACPI in this matter only serves to display your ignorance of > what it is and how it works. Please educate yourself on the > real issues first, thanks. Please take a look at the picture of Warner's laptop which failed to turn on the fans or reduce the CPU speed when the temperature went up, and warped the display and keyboard as it practically melted them into unrecognizability, thanks to the ACPI not working like it was supposed to work, even if you turned off autosuspend. Also, please read the Intel Multiprocessing Specification version 1.4, and note that it says nothing whatsoever about ACPI replacing the MPTable, and that the MPTable is the only documented standard mechanism for implementing obtaining MP information in an Intel Multiprocessing Specification version 1.4 compliant fashion. ACPI may have a cleaner interface, but standards rule when it comes to trying to get software running (and I have yet to see APM turn a laptop into "The Incredible Melting Man": anything that works is better than anything that doesn't). -- Terry From owner-freebsd-smp@FreeBSD.ORG Thu Jul 17 09:16:04 2003 Return-Path: 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 BD11737B401 for ; Thu, 17 Jul 2003 09:16:04 -0700 (PDT) Received: from web3.globalcom.cl (alpha.globalcom.cl [64.76.146.65]) by mx1.FreeBSD.org (Postfix) with SMTP id EE7BC43FA3 for ; Thu, 17 Jul 2003 09:16:00 -0700 (PDT) (envelope-from infogc@globalcom.cl) Received: (qmail 27438 invoked by uid 0); 17 Jul 2003 15:52:09 -0000 Received: from unknown (HELO dns) (64.76.146.66) by 172.22.0.35 with SMTP; 17 Jul 2003 15:52:09 -0000 Message-ID: <005901c34c7b$259adba0$018014ac@globalcom.cl> From: "infogc" To: Date: Thu, 17 Jul 2003 11:50:24 -0400 MIME-Version: 1.0 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 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.1 Subject: Memory X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 17 Jul 2003 16:16:05 -0000 Please: Y need Buy (2) Memories MEM-16BF-53=3D Do you Have? How Many the cost? Thank You Mar=EDa Fern=E1ndez S. Globalcom S.A. Phone: 562-2153176