From owner-freebsd-smp Mon Jan 31 4:31:40 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mercury.is.co.za (mercury.is.co.za [196.4.160.222]) by hub.freebsd.org (Postfix) with ESMTP id 240B614E21 for ; Mon, 31 Jan 2000 04:31:28 -0800 (PST) (envelope-from conradj@is.co.za) Received: from hermwas.is.co.za (hermwas.is.co.za [196.23.0.8]) by mercury.is.co.za (8.9.3/8.9.3) with ESMTP id OAA27110 for ; Mon, 31 Jan 2000 14:31:21 +0200 Received: from tom-jones.is.co.za (tom-jones.is.co.za [196.23.0.193]) by hermwas.is.co.za (8.9.3/8.9.3) with ESMTP id OAA22446 for ; Mon, 31 Jan 2000 14:31:19 +0200 (SAT) Received: by tom-jones.is.co.za (Postfix, from userid 1001) id C68B9325; Mon, 31 Jan 2000 14:31:22 +0200 (SAST) Date: Mon, 31 Jan 2000 14:31:22 +0200 From: Conrad Juleff To: smp@freebsd.org Subject: Dell 2400 and APIC problem Message-ID: <20000131143122.G53753@is.co.za> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary=IS0zKkzwUGydFO0o X-Mailer: Mutt 0.95.4i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org --IS0zKkzwUGydFO0o Content-Type: text/plain; charset=us-ascii I have a Dell 2400 with 2x 500 PIII cpu's. I cannot get an SMP kernel to run on this machine and I believe it has something to do with the Reliance/RCC chipset. It has 2 x IO APIC's and there seems to be a problem with this. I have looked at the SMP page on freebsd.org and it recommends disabling the 2nd IO APIC in the bios. The bios doesnt have this option and the vendor tells me it cannot be disabled. The machine runs fine in UP mode. I have built a kernel with DDB and debugging and stepped through it. There is a problem setting up the APIC's but I cannot save the output since it flashes past and the machine reboots. Without running boot -d the machine hangs and freezes totally after the serial ports are detected(I dont think it can be related to this) and there is nothing I can do except a reset. I have attached the output of mptable and dmesg(in UP mode). What can I do to locate the exact problem and how do I fix it? I have sent through a send-pr but have had no response. Regards Conrad --IS0zKkzwUGydFO0o Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename="mptable.out" =============================================================================== MPTable, version 2.0.15 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000fe710 signature: '_MP_' length: 16 bytes version: 1.4 checksum: 0x91 mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x000f0000 signature: 'PCMP' base table length: 468 version: 1.4 checksum: 0x53 OEM ID: 'DELL ' Product ID: 'POWEREDGE 9B' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 50 local APIC address: 0xfee00000 extended table length: 88 extended table checksum: 246 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 1 0x11 BSP, usable 6 7 3 0x383fbff 0 0x11 AP, usable 6 7 3 0x383fbff -- Bus: Bus ID Type 0 PCI 1 PCI 2 PCI 3 ISA -- I/O APICs: APIC ID Version State Address 2 0x11 usable 0xfec00000 3 0x11 usable 0xfec01000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 3 0 2 0 INT conforms conforms 3 1 2 1 INT conforms conforms 3 3 2 3 INT conforms conforms 3 4 2 4 INT conforms conforms 3 5 2 5 INT conforms conforms 3 6 2 6 INT conforms conforms 3 7 2 7 INT conforms conforms 3 8 2 8 INT conforms conforms 3 9 2 9 INT conforms conforms 3 12 2 12 INT conforms conforms 3 15 2 15 INT conforms conforms 0 8:A 3 0 INT conforms conforms 1 6:A 3 14 INT conforms conforms 1 4:A 3 15 INT conforms conforms 2 6:A 3 12 INT conforms conforms 2 6:C 3 12 INT conforms conforms 2 6:B 3 13 INT conforms conforms 2 6:D 3 13 INT conforms conforms 2 8:A 3 10 INT conforms conforms 2 8:C 3 10 INT conforms conforms 2 8:B 3 11 INT conforms conforms 2 8:D 3 11 INT conforms conforms 2 10:A 3 8 INT conforms conforms 2 10:C 3 8 INT conforms conforms 2 10:B 3 9 INT conforms conforms 2 10:D 3 9 INT conforms conforms 2 12:A 3 6 INT conforms conforms 2 12:C 3 6 INT conforms conforms 2 12:B 3 7 INT conforms conforms 2 12:D 3 7 INT conforms conforms 2 14:A 3 4 INT conforms conforms 2 14:C 3 4 INT conforms conforms 2 14:B 3 5 INT conforms conforms 2 14:D 3 5 INT conforms conforms 0 2:B 3 3 INT conforms conforms 0 2:A 3 15 INT conforms conforms 0 4:A 3 1 INT conforms conforms 0 4:B 3 2 INT conforms conforms 0 4:C 3 1 INT conforms conforms 0 4:D 3 2 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 3 0 255 0 NMI active-hi edge 3 0 255 1 ------------------------------------------------------------------------------- MP Config Extended Table Entries: -- bus ID: 0 address type: I/O address address base: 0xe000 address range: 0x2000 -- bus ID: 0 address type: memory address address base: 0xa0000 address range: 0x20000 -- bus ID: 0 address type: I/O address address base: 0x0 address range: 0x1000 -- bus ID: 0 address type: memory address address base: 0xf7000000 address range: 0x7110000 -- bus ID: 3 bus info: 0x01 parent bus ID: 0 ------------------------------------------------------------------------------- # SMP kernel config file options: # Required: options SMP # Symmetric MultiProcessor Kernel options APIC_IO # Symmetric (APIC) I/O # Optional (built-in defaults will work in most cases): #options NCPU=2 # number of CPUs #options NBUS=4 # number of busses #options NAPIC=2 # number of IO APICs #options NINTR=40 # number of INTs =============================================================================== --IS0zKkzwUGydFO0o Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename="dmesg.out" Copyright (c) 1992-1999 FreeBSD Inc. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 3.3-19991117-STABLE #0: Fri Jan 28 08:54:48 SAST 2000 root@eos.is.co.za:/usr/src/sys/compile/INFEED Timecounter "i8254" frequency 1193182 Hz Timecounter "TSC" frequency 498342864 Hz CPU: Pentium III (498.34-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x673 Stepping = 3 Features=0x383fbff> real memory = 1073741824 (1048576K bytes) config> q avail memory = 1042276352 (1017848K bytes) Preloaded elf kernel "kernel" at 0xc02a9000. Preloaded userconfig_script "/boot/kernel.conf" at 0xc02a909c. Pentium Pro MTRR support enabled Probing for devices on PCI bus 0: chip0: rev 0x05 on pci0.0.0 chip1: rev 0x05 on pci0.0.1 chip2: rev 0x01 on pci0.2.0 fxp0: rev 0x08 int a irq 10 on pci0.8.0 fxp0: Ethernet address 00:c0:4f:01:19:57 vga0: rev 0x7a on pci0.14.0 chip3: rev 0x4f on pci0.15.0 Probing for devices on PCI bus 1: ahc0: rev 0x01 int a irq 14 on pci1.4.0 ahc0: aic7890/91 Wide Channel A, SCSI Id=7, 16/255 SCBs ahc1: rev 0x01 int a irq 14 on pci1.6.0 ahc1: aic7880 Single Channel A, SCSI Id=7, 16/255 SCBs Probing for devices on the ISA bus: sc0 on isa sc0: VGA color <16 virtual consoles, flags=0x0> atkbdc0 at 0x60-0x6f on motherboard atkbd0 irq 1 on isa fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: FIFO enabled, 8 bytes threshold fd0: 1.44MB 3.5in aha0 not found at 0x134 vga0 at 0x3b0-0x3df maddr 0xa0000 msize 131072 on isa npx0 on motherboard npx0: INT 16 interface Waiting 15 seconds for SCSI devices to settle pass3 at ahc0 bus 0 target 6 lun 0 pass3: Fixed Processor SCSI-2 device pass3: 3.300MB/s transfers changing root device to da0s1a cd0 at ahc1 bus 0 target 5 lun 0 cd0: Removable CD-ROM SCSI-2 device cd0: 20.000MB/s transfers (20.000MHz, offset 15) cd0: Attempt to query device size failed: NOT READY, Medium not present da1 at ahc0 bus 0 target 1 lun 0 da1: Fixed Direct Access SCSI-3 device da1: 80.000MB/s transfers (40.000MHz, offset 31, 16bit), Tagged Queueing Enabled da1: 17366MB (35566501 512 byte sectors: 255H 63S/T 2213C) --IS0zKkzwUGydFO0o-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Jan 31 9:10: 3 2000 Delivered-To: freebsd-smp@freebsd.org Received: from newyork.tycosweb.com (newyork.tycosweb.com [209.8.205.4]) by hub.freebsd.org (Postfix) with SMTP id 3F76D14D1C for ; Mon, 31 Jan 2000 09:10:01 -0800 (PST) (envelope-from simon@tycosweb.com) Received: (qmail 10418 invoked from network); 31 Jan 2000 17:10:46 -0000 Received: from ppp27-2.asan.com (HELO stdio) (admin@207.113.83.27) by newyork.tycosweb.com with SMTP; 31 Jan 2000 17:10:46 -0000 From: "Simon" To: "freebsd-smp@FreeBSD.ORG" Date: Mon, 31 Jan 2000 12:14:52 -0500 Reply-To: "Simon" X-Mailer: PMMail 98 Pro (2.01.1600) For Windows 98 (4.10.1998) MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Subject: subscribe Message-Id: <20000131171001.3F76D14D1C@hub.freebsd.org> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org subscribe To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 3:46:39 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mercury.is.co.za (mercury.is.co.za [196.4.160.222]) by builder.freebsd.org (Postfix) with ESMTP id 7A9013DA3; Tue, 1 Feb 2000 03:46:22 -0800 (PST) Received: from hermwas.is.co.za (hermwas.is.co.za [196.23.0.8]) by mercury.is.co.za (8.9.3/8.9.3) with ESMTP id NAA14741; Tue, 1 Feb 2000 13:46:08 +0200 Received: from tom-jones.is.co.za (tom-jones.is.co.za [196.23.0.193]) by hermwas.is.co.za (8.9.3/8.9.3) with ESMTP id NAA17795; Tue, 1 Feb 2000 13:46:08 +0200 (SAT) Received: by tom-jones.is.co.za (Postfix, from userid 1001) id 42E8332D; Tue, 1 Feb 2000 13:46:11 +0200 (SAST) Date: Tue, 1 Feb 2000 13:46:11 +0200 From: Conrad Juleff To: smp@freebsd.org Cc: current@freebsd.org Subject: Re: Dell 2400 and APIC problem Message-ID: <20000201134611.E57702@is.co.za> References: <20000131143122.G53753@is.co.za> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.4i In-Reply-To: <20000131143122.G53753@is.co.za>; from Conrad Juleff on Mon, Jan 31, 2000 at 02:31:22PM +0200 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org After much messing around I have found the point on current where it hangs. It seems to hang during the PnP probe: isa_probe_children: probing PnP devices Is there anyway I can stop this probing of PnP devices? I have tried adding "optons PNPBIOS" to the kernel but it causes the kernel to panic. If I drop into the debugger and type panic, the kernel seems to be stuck in mp_lock. I also cant seem to get the kernel to save core. The bios doesnt have an option to disable PnP. Any ideas on what else I can do? On Mon, Jan 31, 2000 at 02:31:22PM +0200, Conrad Juleff wrote: > I have a Dell 2400 with 2x 500 PIII cpu's. I cannot get an SMP kernel to run > on this machine and I believe it has something to do with the Reliance/RCC > chipset. It has 2 x IO APIC's and there seems to be a problem with this. I > have looked at the SMP page on freebsd.org and it recommends disabling the 2nd > IO APIC in the bios. The bios doesnt have this option and the vendor tells me > it cannot be disabled. The machine runs fine in UP mode. > > I have built a kernel with DDB and debugging and stepped through it. There is > a problem setting up the APIC's but I cannot save the output since it flashes > past and the machine reboots. Without running boot -d the machine hangs and > freezes totally after the serial ports are detected(I dont think it can be > related to this) and there is nothing I can do except a reset. I have attached > the output of mptable and dmesg(in UP mode). > > What can I do to locate the exact problem and how do I fix it? I have sent > through a send-pr but have had no response. > > > Regards > Conrad > > =============================================================================== > > MPTable, version 2.0.15 > > ------------------------------------------------------------------------------- > > MP Floating Pointer Structure: > > location: BIOS > physical address: 0x000fe710 > signature: '_MP_' > length: 16 bytes > version: 1.4 > checksum: 0x91 > mode: Virtual Wire > > ------------------------------------------------------------------------------- > > MP Config Table Header: > > physical address: 0x000f0000 > signature: 'PCMP' > base table length: 468 > version: 1.4 > checksum: 0x53 > OEM ID: 'DELL ' > Product ID: 'POWEREDGE 9B' > OEM table pointer: 0x00000000 > OEM table size: 0 > entry count: 50 > local APIC address: 0xfee00000 > extended table length: 88 > extended table checksum: 246 > > ------------------------------------------------------------------------------- > > MP Config Base Table Entries: > > -- > Processors: APIC ID Version State Family Model Step Flags > 1 0x11 BSP, usable 6 7 3 0x383fbff > 0 0x11 AP, usable 6 7 3 0x383fbff > -- > Bus: Bus ID Type > 0 PCI > 1 PCI > 2 PCI > 3 ISA > -- > I/O APICs: APIC ID Version State Address > 2 0x11 usable 0xfec00000 > 3 0x11 usable 0xfec01000 > -- > I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# > ExtINT active-hi edge 3 0 2 0 > INT conforms conforms 3 1 2 1 > INT conforms conforms 3 3 2 3 > INT conforms conforms 3 4 2 4 > INT conforms conforms 3 5 2 5 > INT conforms conforms 3 6 2 6 > INT conforms conforms 3 7 2 7 > INT conforms conforms 3 8 2 8 > INT conforms conforms 3 9 2 9 > INT conforms conforms 3 12 2 12 > INT conforms conforms 3 15 2 15 > INT conforms conforms 0 8:A 3 0 > INT conforms conforms 1 6:A 3 14 > INT conforms conforms 1 4:A 3 15 > INT conforms conforms 2 6:A 3 12 > INT conforms conforms 2 6:C 3 12 > INT conforms conforms 2 6:B 3 13 > INT conforms conforms 2 6:D 3 13 > INT conforms conforms 2 8:A 3 10 > INT conforms conforms 2 8:C 3 10 > INT conforms conforms 2 8:B 3 11 > INT conforms conforms 2 8:D 3 11 > INT conforms conforms 2 10:A 3 8 > INT conforms conforms 2 10:C 3 8 > INT conforms conforms 2 10:B 3 9 > INT conforms conforms 2 10:D 3 9 > INT conforms conforms 2 12:A 3 6 > INT conforms conforms 2 12:C 3 6 > INT conforms conforms 2 12:B 3 7 > INT conforms conforms 2 12:D 3 7 > INT conforms conforms 2 14:A 3 4 > INT conforms conforms 2 14:C 3 4 > INT conforms conforms 2 14:B 3 5 > INT conforms conforms 2 14:D 3 5 > INT conforms conforms 0 2:B 3 3 > INT conforms conforms 0 2:A 3 15 > INT conforms conforms 0 4:A 3 1 > INT conforms conforms 0 4:B 3 2 > INT conforms conforms 0 4:C 3 1 > INT conforms conforms 0 4:D 3 2 > -- > Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# > ExtINT active-hi edge 3 0 255 0 > NMI active-hi edge 3 0 255 1 > > ------------------------------------------------------------------------------- > > MP Config Extended Table Entries: > > -- > > bus ID: 0 address type: I/O address > address base: 0xe000 > address range: 0x2000 > -- > > bus ID: 0 address type: memory address > address base: 0xa0000 > address range: 0x20000 > -- > > bus ID: 0 address type: I/O address > address base: 0x0 > address range: 0x1000 > -- > > bus ID: 0 address type: memory address > address base: 0xf7000000 > address range: 0x7110000 > -- > > bus ID: 3 bus info: 0x01 parent bus ID: 0 > ------------------------------------------------------------------------------- > > # SMP kernel config file options: > > > # Required: > options SMP # Symmetric MultiProcessor Kernel > options APIC_IO # Symmetric (APIC) I/O > > # Optional (built-in defaults will work in most cases): > #options NCPU=2 # number of CPUs > #options NBUS=4 # number of busses > #options NAPIC=2 # number of IO APICs > #options NINTR=40 # number of INTs > > =============================================================================== > > Copyright (c) 1992-1999 FreeBSD Inc. > Copyright (c) 1982, 1986, 1989, 1991, 1993 > The Regents of the University of California. All rights reserved. > FreeBSD 3.3-19991117-STABLE #0: Fri Jan 28 08:54:48 SAST 2000 > root@eos.is.co.za:/usr/src/sys/compile/INFEED > Timecounter "i8254" frequency 1193182 Hz > Timecounter "TSC" frequency 498342864 Hz > CPU: Pentium III (498.34-MHz 686-class CPU) > Origin = "GenuineIntel" Id = 0x673 Stepping = 3 > Features=0x383fbff> > real memory = 1073741824 (1048576K bytes) > config> q > avail memory = 1042276352 (1017848K bytes) > Preloaded elf kernel "kernel" at 0xc02a9000. > Preloaded userconfig_script "/boot/kernel.conf" at 0xc02a909c. > Pentium Pro MTRR support enabled > Probing for devices on PCI bus 0: > chip0: rev 0x05 on pci0.0.0 > chip1: rev 0x05 on pci0.0.1 > chip2: rev 0x01 on pci0.2.0 > fxp0: rev 0x08 int a irq 10 on pci0.8.0 > fxp0: Ethernet address 00:c0:4f:01:19:57 > vga0: rev 0x7a on pci0.14.0 > chip3: rev 0x4f on pci0.15.0 > Probing for devices on PCI bus 1: > ahc0: rev 0x01 int a irq 14 on pci1.4.0 > ahc0: aic7890/91 Wide Channel A, SCSI Id=7, 16/255 SCBs > ahc1: rev 0x01 int a irq 14 on pci1.6.0 > ahc1: aic7880 Single Channel A, SCSI Id=7, 16/255 SCBs > Probing for devices on the ISA bus: > sc0 on isa > sc0: VGA color <16 virtual consoles, flags=0x0> > atkbdc0 at 0x60-0x6f on motherboard > atkbd0 irq 1 on isa > fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa > fdc0: FIFO enabled, 8 bytes threshold > fd0: 1.44MB 3.5in > aha0 not found at 0x134 > vga0 at 0x3b0-0x3df maddr 0xa0000 msize 131072 on isa > npx0 on motherboard > npx0: INT 16 interface > Waiting 15 seconds for SCSI devices to settle > pass3 at ahc0 bus 0 target 6 lun 0 > pass3: Fixed Processor SCSI-2 device > pass3: 3.300MB/s transfers > changing root device to da0s1a > cd0 at ahc1 bus 0 target 5 lun 0 > cd0: Removable CD-ROM SCSI-2 device > cd0: 20.000MB/s transfers (20.000MHz, offset 15) > cd0: Attempt to query device size failed: NOT READY, Medium not present > da1 at ahc0 bus 0 target 1 lun 0 > da1: Fixed Direct Access SCSI-3 device > da1: 80.000MB/s transfers (40.000MHz, offset 31, 16bit), Tagged Queueing Enabled > da1: 17366MB (35566501 512 byte sectors: 255H 63S/T 2213C) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 3:54:59 2000 Delivered-To: freebsd-smp@freebsd.org Received: from fw.wintelcom.net (ns1.wintelcom.net [209.1.153.20]) by builder.freebsd.org (Postfix) with ESMTP id DB36D3DA0; Tue, 1 Feb 2000 03:54:55 -0800 (PST) Received: (from bright@localhost) by fw.wintelcom.net (8.9.3/8.9.3) id EAA04005; Tue, 1 Feb 2000 04:19:55 -0800 (PST) Date: Tue, 1 Feb 2000 04:19:55 -0800 From: Alfred Perlstein To: Conrad Juleff Cc: smp@FreeBSD.ORG, current@FreeBSD.ORG Subject: Re: Dell 2400 and APIC problem Message-ID: <20000201041954.J24609@fw.wintelcom.net> References: <20000131143122.G53753@is.co.za> <20000201134611.E57702@is.co.za> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000201134611.E57702@is.co.za>; from conradj@is.co.za on Tue, Feb 01, 2000 at 01:46:11PM +0200 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org * Conrad Juleff [000201 04:13] wrote: > After much messing around I have found the point on current where it hangs. It > seems to hang during the PnP probe: > > isa_probe_children: probing PnP devices > > Is there anyway I can stop this probing of PnP devices? I have tried adding > "optons PNPBIOS" to the kernel but it causes the kernel to panic. If I drop > into the debugger and type panic, the kernel seems to be stuck in mp_lock. I > also cant seem to get the kernel to save core. The bios doesnt have an option > to disable PnP. > > Any ideas on what else I can do? > > Hmm, it seems a lot of the devices in GENERIC have flags to disable PnP probes (see LINT), maybe you can try a custom kernel to install, or try using the kernel config to set the appropriate flags on each device to disable pnp probing. -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 3:56:34 2000 Delivered-To: freebsd-smp@freebsd.org Received: from fw.wintelcom.net (ns1.wintelcom.net [209.1.153.20]) by builder.freebsd.org (Postfix) with ESMTP id CABD73DA9 for ; Tue, 1 Feb 2000 03:56:31 -0800 (PST) Received: (from bright@localhost) by fw.wintelcom.net (8.9.3/8.9.3) id EAA04098; Tue, 1 Feb 2000 04:21:41 -0800 (PST) Date: Tue, 1 Feb 2000 04:21:41 -0800 From: Alfred Perlstein To: Conrad Juleff Cc: smp@FreeBSD.ORG Subject: Re: Dell 2400 and APIC problem Message-ID: <20000201042141.K24609@fw.wintelcom.net> References: <20000131143122.G53753@is.co.za> <20000201134611.E57702@is.co.za> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000201134611.E57702@is.co.za>; from conradj@is.co.za on Tue, Feb 01, 2000 at 01:46:11PM +0200 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org * Conrad Juleff [000201 04:14] wrote: > After much messing around I have found the point on current where it hangs. It > seems to hang during the PnP probe: > > isa_probe_children: probing PnP devices > > Is there anyway I can stop this probing of PnP devices? I have tried adding > "optons PNPBIOS" to the kernel but it causes the kernel to panic. If I drop > into the debugger and type panic, the kernel seems to be stuck in mp_lock. I > also cant seem to get the kernel to save core. The bios doesnt have an option > to disable PnP. > > Any ideas on what else I can do? err, one more thing, this is a 3.3-stable snapshot it seems, have you tried 3.4-stable (a more recent snapshot) or maybe even -current? -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 4:10:44 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mercury.is.co.za (mercury.is.co.za [196.4.160.222]) by builder.freebsd.org (Postfix) with ESMTP id 2A6E43DBD for ; Tue, 1 Feb 2000 04:10:37 -0800 (PST) Received: from hermwas.is.co.za (hermwas.is.co.za [196.23.0.8]) by mercury.is.co.za (8.9.3/8.9.3) with ESMTP id OAA20170; Tue, 1 Feb 2000 14:09:39 +0200 Received: from tom-jones.is.co.za (tom-jones.is.co.za [196.23.0.193]) by hermwas.is.co.za (8.9.3/8.9.3) with ESMTP id OAA18189; Tue, 1 Feb 2000 14:09:38 +0200 (SAT) Received: by tom-jones.is.co.za (Postfix, from userid 1001) id 5D47D32D; Tue, 1 Feb 2000 14:09:41 +0200 (SAST) Date: Tue, 1 Feb 2000 14:09:41 +0200 From: Conrad Juleff To: Alfred Perlstein Cc: smp@FreeBSD.ORG Subject: Re: Dell 2400 and APIC problem Message-ID: <20000201140941.F57702@is.co.za> References: <20000131143122.G53753@is.co.za> <20000201134611.E57702@is.co.za> <20000201042141.K24609@fw.wintelcom.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.4i In-Reply-To: <20000201042141.K24609@fw.wintelcom.net>; from Alfred Perlstein on Tue, Feb 01, 2000 at 04:21:41AM -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, Feb 01, 2000 at 04:21:41AM -0800, Alfred Perlstein wrote: > err, one more thing, this is a 3.3-stable snapshot it seems, have you > tried 3.4-stable (a more recent snapshot) or maybe even -current? I have tried 3.4-stable and a number of recent -current's but they all hang on boot. I am trying to hardwire those adresses in the kernel config as you suggested and I will see what happens. Conrad To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 5:26:16 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mercury.is.co.za (mercury.is.co.za [196.4.160.222]) by builder.freebsd.org (Postfix) with ESMTP id 9DF6A3DAE; Tue, 1 Feb 2000 05:26:04 -0800 (PST) Received: from hermwas.is.co.za (hermwas.is.co.za [196.23.0.8]) by mercury.is.co.za (8.9.3/8.9.3) with ESMTP id PAA10347; Tue, 1 Feb 2000 15:25:50 +0200 Received: from tom-jones.is.co.za (tom-jones.is.co.za [196.23.0.193]) by hermwas.is.co.za (8.9.3/8.9.3) with ESMTP id PAA19827; Tue, 1 Feb 2000 15:25:49 +0200 (SAT) Received: by tom-jones.is.co.za (Postfix, from userid 1001) id BF18B32D; Tue, 1 Feb 2000 15:25:52 +0200 (SAST) Date: Tue, 1 Feb 2000 15:25:52 +0200 From: Conrad Juleff Cc: smp@FreeBSD.ORG, current@FreeBSD.ORG Subject: Re: Dell 2400 and APIC problem Message-ID: <20000201152552.G57702@is.co.za> References: <20000131143122.G53753@is.co.za> <20000201134611.E57702@is.co.za> <20000201041954.J24609@fw.wintelcom.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.4i In-Reply-To: <20000201041954.J24609@fw.wintelcom.net>; from Alfred Perlstein on Tue, Feb 01, 2000 at 04:19:55AM -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, Feb 01, 2000 at 04:19:55AM -0800, Alfred Perlstein wrote: > > Is there anyway I can stop this probing of PnP devices? I have tried adding > > "optons PNPBIOS" to the kernel but it causes the kernel to panic. If I drop > > into the debugger and type panic, the kernel seems to be stuck in mp_lock. I > > also cant seem to get the kernel to save core. The bios doesnt have an option > > to disable PnP. > > > > Any ideas on what else I can do? > > > > > > Hmm, it seems a lot of the devices in GENERIC have flags to disable > PnP probes (see LINT), maybe you can try a custom kernel to install, > or try using the kernel config to set the appropriate flags on each > device to disable pnp probing. I have tried doing this on -current but it doesnt make any difference. I have also removed every non-essential item out of the kernel. I am using boot -dv with "options DIAGNOSTIC" in the kernel and it doesnt tell me what the PnP probe is hanging on. Any other suggestions? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 9: 9:23 2000 Delivered-To: freebsd-smp@freebsd.org Received: from lor.watermarkgroup.com (lor.watermarkgroup.com [207.202.73.33]) by builder.freebsd.org (Postfix) with ESMTP id 2D3F13F17; Tue, 1 Feb 2000 09:09:09 -0800 (PST) Received: (from luoqi@localhost) by lor.watermarkgroup.com (8.8.8/8.8.8) id MAA29645; Tue, 1 Feb 2000 12:08:38 -0500 (EST) (envelope-from luoqi) Date: Tue, 1 Feb 2000 12:08:38 -0500 (EST) From: Luoqi Chen Message-Id: <200002011708.MAA29645@lor.watermarkgroup.com> To: conradj@is.co.za Subject: Re: Dell 2400 and APIC problem Cc: current@FreeBSD.ORG, smp@FreeBSD.ORG Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I have tried doing this on -current but it doesnt make any difference. I have > also removed every non-essential item out of the kernel. > I am using boot -dv with "options DIAGNOSTIC" in the kernel and it doesnt tell > me what the PnP probe is hanging on. > > Any other suggestions? > I don't think it's pnp probing. You don't seem to have any pnp devices at all from your dmesg output. Could you do a 'pnpinfo' and verify if it is indeed the case? Isa bus is the last bus probed, and immediately (well, almost) follows that configure hooks with interrupt enabled. That is also the first time BSP gives up mplock, and the APs have a chance to initialize themselves. I believe that is where it hangs. Just to make sure, do you have "options NAPIC 2" in your kernel config? The default is 1, so you absolutely need this line for things to work. -lq To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 17: 5:58 2000 Delivered-To: freebsd-smp@freebsd.org Received: from web112.yahoomail.com (web112.yahoomail.com [205.180.60.82]) by builder.freebsd.org (Postfix) with SMTP id 966A13D70 for ; Tue, 1 Feb 2000 17:05:56 -0800 (PST) Received: (qmail 16865 invoked by uid 60001); 2 Feb 2000 01:05:50 -0000 Message-ID: <20000202010550.16864.qmail@web112.yahoomail.com> Received: from [203.127.104.35] by web112.yahoomail.com; Tue, 01 Feb 2000 17:05:50 PST Date: Tue, 1 Feb 2000 17:05:50 -0800 (PST) From: Chee Wei Ng Subject: MPrellock_edx To: freebsd-smp@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, I would like to know why we need lock addl $0,0(%esp) /* see note above */ for serialization. Could you show me an example for MP case where it may cause trouble if the above lines are not added in it? Because I didn't see how instruction execution out of order come into the picture since before any processors enter the Critical Section, it has to acquire the mplock first, and acquire the mplock, you must 'LOCK' the bus cycle to serialize the mplock flag to be read-modify-write, so I thought here will do all the serialization as required. Unless, it could be something that may needs to serialize for access before this. Thanks. Rgds, NgCW __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 17:24:44 2000 Delivered-To: freebsd-smp@freebsd.org Received: from fw.wintelcom.net (ns1.wintelcom.net [209.1.153.20]) by builder.freebsd.org (Postfix) with ESMTP id 4A6CE3FDC for ; Tue, 1 Feb 2000 17:24:42 -0800 (PST) Received: (from bright@localhost) by fw.wintelcom.net (8.9.3/8.9.3) id RAA23749; Tue, 1 Feb 2000 17:50:09 -0800 (PST) Date: Tue, 1 Feb 2000 17:50:08 -0800 From: Alfred Perlstein To: Chee Wei Ng Cc: freebsd-smp@FreeBSD.ORG Subject: Re: MPrellock_edx Message-ID: <20000201175008.F24609@fw.wintelcom.net> References: <20000202010550.16864.qmail@web112.yahoomail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000202010550.16864.qmail@web112.yahoomail.com>; from scip7050@yahoo.com on Tue, Feb 01, 2000 at 05:05:50PM -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org * Chee Wei Ng [000201 17:31] wrote: > Hi, > > I would like to know why we need > lock > addl $0,0(%esp) /* see note above */ > for serialization. > > Could you show me an example for MP case where it may cause trouble if the > above lines are not added in it? > > Because I didn't see how instruction execution out of order come into the > picture since before any processors enter the Critical Section, it has to > acquire the mplock first, and acquire the mplock, you must 'LOCK' the bus cycle > to serialize the mplock flag to be read-modify-write, so I thought here will do > all the serialization as required. Unless, it could be something that may needs > to serialize for access before this. It's to ensure that memory ops scheduled _before_ the lock is released have been completed before the lock is actually released. Otherwise out of order memory writes can occur corrupting the state of protected variables. Imagine if a CPU releases a lock then a previously sheduled write on the _same_ cpu goes in several cycles after another processor aquires the lock. Since we aren't using a locked cycle to release the lock, we must _at least_ insert a barrier instruction to force correct ordering. -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 18:30:54 2000 Delivered-To: freebsd-smp@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [216.240.41.2]) by builder.freebsd.org (Postfix) with ESMTP id 7C91B4012 for ; Tue, 1 Feb 2000 18:30:51 -0800 (PST) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id SAA93312; Tue, 1 Feb 2000 18:30:35 -0800 (PST) (envelope-from dillon) Date: Tue, 1 Feb 2000 18:30:35 -0800 (PST) From: Matthew Dillon Message-Id: <200002020230.SAA93312@apollo.backplane.com> To: Alfred Perlstein Cc: Chee Wei Ng , freebsd-smp@FreeBSD.ORG Subject: Re: MPrellock_edx References: <20000202010550.16864.qmail@web112.yahoomail.com> <20000201175008.F24609@fw.wintelcom.net> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I discussed this issue with Linus after someone from the linux kernel group brought it up. Basically we have adopted the same thing that linux uses. Under Intel, the following is true: * Writes are buffered but are committed to memory in the same order they were issued. Thus write<->write conflicts are not an issue. * Speculative reads may occur, but they occur from main memory into the L1 cache and thus still adhere to L1 cache protocols. Thus speculative reads are not an issue. * The cpu may reorder non-conflicting reads. A non-conflicting read may be reordered from *before* a write to *after* a write, or from *after* a write to *before* a write. This is an issue. This is the ONLY issue with intel hardware. The purpose of the locked instruction is to prevent any possibility of reads being reordered from to after the MP lock is released. There has been a huge amount of misinformation on the issue. I looked at both the linux kernel and the Freebsd mail archives and 90% of the messages posted entertaining one opinion or another were just plain wrong (and Linus agrees with me). We don't know whether the read reordering issue is real or not. We do know that none of the other issues brought up were real. We aren't taking any chances with the read re-ordering issue and so we have a locked instruction. The reason we use 0(%esp) for the locked memory address is because there is nearly a 100% chance that that address is already in our processor's L1 cache *AND* that we have (via the hardware cache protocol) exclusive ownership of the address, thus minimizing the cost of running the instruction. The reason we do not use a locked instruction to actually release the MP lock is because: (A) we don't have to, writes are ordered and we do not care if reads are reordered to occur before the write releasing the lock is committed. (B) because if there is another processor trying for the lock we may not have exclusive ownership of the lock address in our L1 cache (the other cpu might), costing us a huge stall due to the way the hardware cache coherency protocol work. -Matt Matthew Dillon : :* Chee Wei Ng [000201 17:31] wrote: :> Hi, :> :> I would like to know why we need :> lock :> addl $0,0(%esp) /* see note above */ :> for serialization. :> :> Could you show me an example for MP case where it may cause trouble if the :> above lines are not added in it? :> :> Because I didn't see how instruction execution out of order come into the :> picture since before any processors enter the Critical Section, it has to :> acquire the mplock first, and acquire the mplock, you must 'LOCK' the bus cycle :> to serialize the mplock flag to be read-modify-write, so I thought here will do :> all the serialization as required. Unless, it could be something that may needs :> to serialize for access before this. : :It's to ensure that memory ops scheduled _before_ the lock is released :have been completed before the lock is actually released. : :Otherwise out of order memory writes can occur corrupting the state :of protected variables. : :Imagine if a CPU releases a lock then a previously sheduled write on the :_same_ cpu goes in several cycles after another processor aquires the :lock. : :Since we aren't using a locked cycle to release the lock, we must _at least_ :insert a barrier instruction to force correct ordering. : :-Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 18:55:39 2000 Delivered-To: freebsd-smp@freebsd.org Received: from fw.wintelcom.net (ns1.wintelcom.net [209.1.153.20]) by builder.freebsd.org (Postfix) with ESMTP id 4F2154018 for ; Tue, 1 Feb 2000 18:55:38 -0800 (PST) Received: (from bright@localhost) by fw.wintelcom.net (8.9.3/8.9.3) id TAA25922; Tue, 1 Feb 2000 19:21:07 -0800 (PST) Date: Tue, 1 Feb 2000 19:21:07 -0800 From: Alfred Perlstein To: Matthew Dillon Cc: Chee Wei Ng , freebsd-smp@FreeBSD.ORG Subject: Re: MPrellock_edx Message-ID: <20000201192107.I24609@fw.wintelcom.net> References: <20000202010550.16864.qmail@web112.yahoomail.com> <20000201175008.F24609@fw.wintelcom.net> <200002020230.SAA93312@apollo.backplane.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <200002020230.SAA93312@apollo.backplane.com>; from dillon@apollo.backplane.com on Tue, Feb 01, 2000 at 06:30:35PM -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org * Matthew Dillon [000201 18:56] wrote: > I discussed this issue with Linus after someone from the linux kernel > group brought it up. Basically we have adopted the same thing that linux > uses. > > Under Intel, the following is true: > > * Writes are buffered but are committed to memory in the same > order they were issued. Thus write<->write conflicts are not > an issue. > > * Speculative reads may occur, but they occur from main memory > into the L1 cache and thus still adhere to L1 cache protocols. > Thus speculative reads are not an issue. > > * The cpu may reorder non-conflicting reads. A non-conflicting > read may be reordered from *before* a write to *after* a write, > or from *after* a write to *before* a write. > > This is an issue. This is the ONLY issue with intel hardware. > > The purpose of the locked instruction is to prevent any possibility > of reads being reordered from to after the MP lock is released. ^ before ? So instead of releasing the lock and then writing the the locked object, you could read from the locked object, release the lock then the read actually happens after the lock release. You're not potentially spamming someone else's critical section, but actually possibly extending your own critical section beyond the point where you've released the lock violating your own critical section? -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 18:59: 3 2000 Delivered-To: freebsd-smp@freebsd.org Received: from fw.wintelcom.net (ns1.wintelcom.net [209.1.153.20]) by builder.freebsd.org (Postfix) with ESMTP id 5A7D64022 for ; Tue, 1 Feb 2000 18:59:01 -0800 (PST) Received: (from bright@localhost) by fw.wintelcom.net (8.9.3/8.9.3) id TAA25953; Tue, 1 Feb 2000 19:24:32 -0800 (PST) Date: Tue, 1 Feb 2000 19:24:32 -0800 From: Alfred Perlstein To: Chee Wei Ng Cc: smp@freebsd.org Subject: Re: MPrellock_edx Message-ID: <20000201192431.J24609@fw.wintelcom.net> References: <20000202022404.5985.qmail@web123.yahoomail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000202022404.5985.qmail@web123.yahoomail.com>; from scip7050@yahoo.com on Tue, Feb 01, 2000 at 06:24:04PM -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org * Chee Wei Ng [000201 19:00] wrote: > Hi Alfred, > > --- Alfred Perlstein wrote: > > * Chee Wei Ng [000201 17:31] wrote: > > > Hi, > > > > > > I would like to know why we need > > > lock > > > addl $0,0(%esp) /* see note above */ > > > for serialization. > > > > > > Could you show me an example for MP case where it may cause trouble if the > > > above lines are not added in it? > > > > > > Because I didn't see how instruction execution out of order come into the > > > picture since before any processors enter the Critical Section, it has to > > > acquire the mplock first, and acquire the mplock, you must 'LOCK' the bus > > cycle > > > to serialize the mplock flag to be read-modify-write, so I thought here > > will do > > > all the serialization as required. Unless, it could be something that may > > needs > > > to serialize for access before this. > > > > It's to ensure that memory ops scheduled _before_ the lock is released > > have been completed before the lock is actually released. > > > > Otherwise out of order memory writes can occur corrupting the state > > of protected variables. > > > > Imagine if a CPU releases a lock then a previously sheduled write on the > > _same_ cpu goes in several cycles after another processor aquires the > > lock. > > > > Since we aren't using a locked cycle to release the lock, we must _at least_ > > insert a barrier instruction to force correct ordering. > > > > -Alfred > > > > In this case, what you mean is that in order to update all the modified memory > for previous locked Critical Section, the barrier must be inserted for that > processor, right? > > For example, > > Initial value of A = 0xdeadc0de; > > Processor 1 Processor 2 > ----------- ----------- > MPgetlock_edx(mplock); > A = 0; > MPrellock_edx(mplock); > MPgetlock(mplock); > read A; > MPrellock(mplock); > > Suppose we do not have barrier instruction at MPrellock. > According to your explanation: > 1. The 'LOCK' instruction executed in MPgetlock at processor 2 after processor > 1 > released the mplock will not serialize the value of A, so when processor 2 got > the mplock and read A, it is possible to read 0xdeadc0de instead of 0. > 2. The write of A and mplock will not be in order, i.e. it could be > mplock == 0 and A == 0xdeadc0de, > after processor 1 released mplock and processor 2 wanted to acquire the mplock. > > The processor 1 might have follow the write order but the processor 2 will see > two values in out of order way. i.e. mplock will be updated before A. > > That's why processor 2 thought that it has the mplock and it's safe to read A, > but infact it is not. > > Is this what you mean? I'm wrong, see Matt Dillon's answer. -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 20:19:36 2000 Delivered-To: freebsd-smp@freebsd.org Received: from web117.yahoomail.com (web117.yahoomail.com [205.180.60.91]) by builder.freebsd.org (Postfix) with SMTP id 3DD6F3DBA for ; Tue, 1 Feb 2000 20:19:34 -0800 (PST) Received: (qmail 4104 invoked by uid 60001); 2 Feb 2000 04:08:33 -0000 Message-ID: <20000202040833.4103.qmail@web117.yahoomail.com> Received: from [203.127.104.35] by web117.yahoomail.com; Tue, 01 Feb 2000 20:08:33 PST Date: Tue, 1 Feb 2000 20:08:33 -0800 (PST) From: Chee Wei Ng Subject: Re: MPrellock_edx To: Matthew Dillon Cc: freebsd-smp@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org --- Matthew Dillon wrote: > I discussed this issue with Linus after someone from the linux kernel > group brought it up. Basically we have adopted the same thing that linux > uses. > > Under Intel, the following is true: > > * Writes are buffered but are committed to memory in the same > order they were issued. Thus write<->write conflicts are not > an issue. > > * Speculative reads may occur, but they occur from main memory > into the L1 cache and thus still adhere to L1 cache protocols. > Thus speculative reads are not an issue. > > * The cpu may reorder non-conflicting reads. A non-conflicting > read may be reordered from *before* a write to *after* a write, > or from *after* a write to *before* a write. > > This is an issue. This is the ONLY issue with intel hardware. > > The purpose of the locked instruction is to prevent any possibility > of reads being reordered from to after the MP lock is released. OK! So the issue here is non-conflict read re-ordering. But I still don't see why there is an impact on corrupting the CS. (Maybe I am too stupid). Further illustration with an example will be grateful. If the above issue is as it is, I understand why you used the below method to resolve the above issue. > > There has been a huge amount of misinformation on the issue. I looked > at both the linux kernel and the Freebsd mail archives and 90% of > the messages posted entertaining one opinion or another were just plain > wrong (and Linus agrees with me). We don't know whether the read > reordering issue is real or not. We do know that none of the other > issues brought up were real. > > We aren't taking any chances with the read re-ordering issue and so > we have a locked instruction. > The reason we use 0(%esp) for the locked memory address is because > there is nearly a 100% chance that that address is already in our > processor's L1 cache *AND* that we have (via the hardware cache > protocol) exclusive ownership of the address, thus minimizing the cost > of running the instruction. > > The reason we do not use a locked instruction to actually release the > MP lock is because: > > (A) we don't have to, writes are ordered and we do not care if > reads are reordered to occur before the write releasing the > lock is committed. > > (B) because if there is another processor trying for the lock we > may not have exclusive ownership of the lock address in our L1 > cache (the other cpu might), costing us a huge stall due to > the way the hardware cache coherency protocol work. > > -Matt > Matthew Dillon > > > Thanks. Rgds, NgCW __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 20:22:58 2000 Delivered-To: freebsd-smp@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [216.240.41.2]) by builder.freebsd.org (Postfix) with ESMTP id 5A8693DBA for ; Tue, 1 Feb 2000 20:22:54 -0800 (PST) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id UAA93953; Tue, 1 Feb 2000 20:22:47 -0800 (PST) (envelope-from dillon) Date: Tue, 1 Feb 2000 20:22:47 -0800 (PST) From: Matthew Dillon Message-Id: <200002020422.UAA93953@apollo.backplane.com> To: Alfred Perlstein Cc: Chee Wei Ng , freebsd-smp@FreeBSD.ORG Subject: Re: MPrellock_edx References: <20000202010550.16864.qmail@web112.yahoomail.com> <20000201175008.F24609@fw.wintelcom.net> <200002020230.SAA93312@apollo.backplane.com> <20000201192107.I24609@fw.wintelcom.net> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org :> * The cpu may reorder non-conflicting reads. A non-conflicting :> read may be reordered from *before* a write to *after* a write, :> or from *after* a write to *before* a write. :> :> This is an issue. This is the ONLY issue with intel hardware. :> :> The purpose of the locked instruction is to prevent any possibility :> of reads being reordered from to after the MP lock is released. : ^ : before ? : :So instead of releasing the lock and then writing the the locked object, :you could read from the locked object, release the lock then the read :actually happens after the lock release. Right. :You're not potentially spamming someone else's critical section, but actually :possibly extending your own critical section beyond the point where you've :released the lock violating your own critical section? : :-Alfred Right. For example: MP lock read critical data MP unlock use the critical data that you read Pentiums have long pipelines and use a duel-issue architecture. The cpu is allowed to reorder non-conflicting instructions so the execution of this: read instruction write instruction (release MP, not using a locked instruction) use data from read instruction May be reordered by the cpu into this: write instruction (release MP, not using a locked instruction) read instruction use data from read instruction The instruction reordering is *NOT* smp-aware if you do not use a locked instruction. The reordering can occur if you read from an address that does not conflict with the write's address. By using a dummy locked instruction, we prevent the pentium from reordering any instructions occuring before the dummy locked instruction to after: read instruction DUMMY LOCKED INSTRUCTION write instruction (release MP lock, not using a locked instruction) Now, obviously, the instruction reordering is heavily dependant on the pipeline and issue architecture. I am 99.9% sure that we run enough instructions in our MP unlock code such that no read instructions could possibly be reordered beyond our write instruction. But I am 0.1% unsure. We just don't know the state of the pipeline upon entry to the MPUnlock code. Plus future cpu's may use a different pipeline/issue architecture, so we cannot count on the current reordering mechanisms to stay the same (remember the 386->486 transition? Lots of people were using pipeline tricks for self modifying code and got screwed because they were violating the letter of the spec). Thus we have that dummy locked instruction sitting there. -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 20:33:31 2000 Delivered-To: freebsd-smp@freebsd.org Received: from web117.yahoomail.com (web117.yahoomail.com [205.180.60.91]) by builder.freebsd.org (Postfix) with SMTP id 482C23DC4 for ; Tue, 1 Feb 2000 20:33:24 -0800 (PST) Received: (qmail 8846 invoked by uid 60001); 2 Feb 2000 04:28:29 -0000 Message-ID: <20000202042829.8845.qmail@web117.yahoomail.com> Received: from [203.127.104.35] by web117.yahoomail.com; Tue, 01 Feb 2000 20:28:29 PST Date: Tue, 1 Feb 2000 20:28:29 -0800 (PST) From: Chee Wei Ng Subject: Re: MPrellock_edx To: freebsd-smp@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org --- Matthew Dillon wrote: > I discussed this issue with Linus after someone from the linux kernel > group brought it up. Basically we have adopted the same thing that linux > uses. > > Under Intel, the following is true: > > * Writes are buffered but are committed to memory in the same > order they were issued. Thus write<->write conflicts are not > an issue. > > * Speculative reads may occur, but they occur from main memory > into the L1 cache and thus still adhere to L1 cache protocols. > Thus speculative reads are not an issue. > > * The cpu may reorder non-conflicting reads. A non-conflicting > read may be reordered from *before* a write to *after* a write, > or from *after* a write to *before* a write. > > This is an issue. This is the ONLY issue with intel hardware. > > The purpose of the locked instruction is to prevent any possibility > of reads being reordered from to after the MP lock is released. > So, the issue is non-conflicting re-ordered read. But, I still don't see why this issue has impact on the CS. Further illustration with an example will be grateful. The following solution will be cleared if the above issue is really an issue. > There has been a huge amount of misinformation on the issue. I looked > at both the linux kernel and the Freebsd mail archives and 90% of > the messages posted entertaining one opinion or another were just plain > wrong (and Linus agrees with me). We don't know whether the read > reordering issue is real or not. We do know that none of the other > issues brought up were real. > > We aren't taking any chances with the read re-ordering issue and so > we have a locked instruction. > > The reason we use 0(%esp) for the locked memory address is because > there is nearly a 100% chance that that address is already in our > processor's L1 cache *AND* that we have (via the hardware cache > protocol) exclusive ownership of the address, thus minimizing the cost > of running the instruction. > > The reason we do not use a locked instruction to actually release the > MP lock is because: > > (A) we don't have to, writes are ordered and we do not care if > reads are reordered to occur before the write releasing the > lock is committed. > > (B) because if there is another processor trying for the lock we > may not have exclusive ownership of the lock address in our L1 > cache (the other cpu might), costing us a huge stall due to > the way the hardware cache coherency protocol work. > > -Matt > Matthew Dillon > > NgCW __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 1 22:16:59 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mercury.is.co.za (mercury.is.co.za [196.4.160.222]) by builder.freebsd.org (Postfix) with ESMTP id 32A893E1F; Tue, 1 Feb 2000 22:16:51 -0800 (PST) Received: from hermwas.is.co.za (hermwas.is.co.za [196.23.0.8]) by mercury.is.co.za (8.9.3/8.9.3) with ESMTP id IAA21035; Wed, 2 Feb 2000 08:16:37 +0200 Received: from tom-jones.is.co.za (tom-jones.is.co.za [196.23.0.193]) by hermwas.is.co.za (8.9.3/8.9.3) with ESMTP id IAA01613; Wed, 2 Feb 2000 08:16:36 +0200 (SAT) Received: by tom-jones.is.co.za (Postfix, from userid 1001) id F28CE32D; Wed, 2 Feb 2000 08:16:39 +0200 (SAST) Date: Wed, 2 Feb 2000 08:16:39 +0200 From: Conrad Juleff To: Luoqi Chen Cc: current@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: Dell 2400 and APIC problem Message-ID: <20000202081639.C61121@is.co.za> References: <200002011708.MAA29645@lor.watermarkgroup.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.4i In-Reply-To: <200002011708.MAA29645@lor.watermarkgroup.com>; from Luoqi Chen on Tue, Feb 01, 2000 at 12:08:38PM -0500 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, Feb 01, 2000 at 12:08:38PM -0500, Luoqi Chen wrote: > > I am using boot -dv with "options DIAGNOSTIC" in the kernel and it doesnt > > tell me what the PnP probe is hanging on. > > > > Any other suggestions? > > > I don't think it's pnp probing. You don't seem to have any pnp devices at > all from your dmesg output. Could you do a 'pnpinfo' and verify if it is > indeed the case? You are correct: # pnpinfo Checking for Plug-n-Play devices... No Plug-n-Play devices were found > Isa bus is the last bus probed, and immediately (well, almost) follows that > configure hooks with interrupt enabled. That is also the first time BSP > gives up mplock, and the APs have a chance to initialize themselves. I > believe that is where it hangs. This looks like the problem, thanks for your help. Do you know how I can narrow this down and perhaps get it working? > Just to make sure, do you have "options NAPIC 2" in your kernel config? The > default is 1, so you absolutely need this line for things to work. Yes I have "options NAPIC 2" in my kernel config, the kernel complains and halts if I have it set to 1. -Conrad To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 8 2: 1:46 2000 Delivered-To: freebsd-smp@freebsd.org Received: from berk.mail.netforce.net (berk.mail.netforce.net [195.58.64.236]) by builder.freebsd.org (Postfix) with ESMTP id 06358419C; Tue, 8 Feb 2000 02:01:40 -0800 (PST) Received: from bagpuss.i.netforce.net (tarbuck.netforce.net [195.58.64.34]) by berk.mail.netforce.net (8.8.8/8.8.7) with ESMTP id KAA12083; Tue, 8 Feb 2000 10:36:35 GMT Date: Tue, 8 Feb 2000 10:08:19 +0000 (GMT) From: James Holtom Reply-To: James Holtom To: freebsd-questions@freebsd.org, freebsd-smp@freebsd.org Subject: (SMP) Celerons + ABIT BP6 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Does anyone have any current [preferably good] experience using FreeBSD 3.4 [Stable] on a twin-celeron ABIT BP6 board? I seem to recall seeing bad reports about stability and so on, but with the mailing list archives dead ATM... Any info gratefully received... Regards, James To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 8 2:21:28 2000 Delivered-To: freebsd-smp@freebsd.org Received: from overcee.netplex.com.au (overcee.netplex.com.au [202.12.86.7]) by builder.freebsd.org (Postfix) with ESMTP id 4FE8F4046; Tue, 8 Feb 2000 02:21:10 -0800 (PST) Received: from netplex.com.au (localhost [127.0.0.1]) by overcee.netplex.com.au (Postfix) with ESMTP id 7457D1CD7; Tue, 8 Feb 2000 18:21:56 +0800 (WST) (envelope-from peter@netplex.com.au) X-Mailer: exmh version 2.1.1 10/15/1999 To: James Holtom Cc: freebsd-questions@freebsd.org, freebsd-smp@freebsd.org Subject: Re: (SMP) Celerons + ABIT BP6 In-Reply-To: Message from James Holtom of "Tue, 08 Feb 2000 10:08:19 GMT." Date: Tue, 08 Feb 2000 18:21:56 +0800 From: Peter Wemm Message-Id: <20000208102156.7457D1CD7@overcee.netplex.com.au> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org James Holtom wrote: > Does anyone have any current [preferably good] experience using FreeBSD > 3.4 [Stable] on a twin-celeron ABIT BP6 board? > > I seem to recall seeing bad reports about stability and so on, but with > the mailing list archives dead ATM... I'm happily running 4.0-CURRENT on an ABIT BP6 w/ 2 celeron 300's with monster cooling fans and running at 450MHz. Apart from it overheating yesterday (there are 6 high-rpm drives in it) when I closed the case for the first time, I had some Wierd Stuff(TM). Taking the cover back off solved it. This isn't suprising since it's very much in the "Don't do that" department and having all the extra drives in a confined space and being in the middle of summer here was asking for trouble (and I got it). Your mileage may vary, don't try overclocking at the very least until you have verified it as stable and no major problems are present. Trying to debug something with malfunctioning overheating cpus isn't much fun. Better still, don't overclock it at all if you value your sanity. In particular do not send in problem reports unless it's reproduceable without overclocking. Beware, the 3.x code does not support the UDMA66 connectors. You can only use the UDMA33 IDE connectors. You have to be running the absolute very latest -current to use the UDMA66 controller. FreeBSD 4.0-CURRENT #92: Sat Feb 5 22:57:37 WST 2000 Timecounter "i8254" frequency 1193264 Hz CPU: Pentium II/Pentium II Xeon/Celeron (451.05-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x665 Stepping = 5 Features=0x183fbff real memory = 67043328 (65472K bytes) avail memory = 60948480 (59520K bytes) Programming 24 pins in IOAPIC #0 IOAPIC #0 intpin 2 -> irq 0 FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 0, version: 0x00040011, at 0xfee00000 cpu1 (AP): apic id: 1, version: 0x00040011, at 0xfee00000 io0 (APIC): apic id: 2, version: 0x00170011, at 0xfec00000 Preloaded elf kernel "kernel" at 0xc03e6000. ... ata-pci5: port 0xe000-0xe0ff,0xdc00-0xdc03,0xd800-0xd807 irq 18 at device 19.0 on pci0 ata6 at 0xd800 irq -1 on ata-pci5 ata-pci6: port 0xec00-0xecff,0xe800-0xe803,0xe400-0xe407 irq 18 at device 19.1 on pci0 ata7 at 0xe400 irq -1 on ata-pci6 ... APIC_IO: Testing 8254 interrupt delivery APIC_IO: routing 8254 via IOAPIC #0 intpin 2 SMP: AP CPU #1 Launched! ... ad0: 9787MB [19885/16/63] at ata2-master using UDMA33 ad1: 6149MB [13328/15/63] at ata3-master using UDMA33 ad2: 6149MB [13328/15/63] at ata4-master using UDMA33 ad3: 6149MB [13328/15/63] at ata5-master using UDMA33 ad4: 6149MB [13328/15/63] at ata6-master using UDMA33 ad5: 6149MB [13328/15/63] at ata7-master using UDMA33 Cheers, -Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 8 2:24:37 2000 Delivered-To: freebsd-smp@freebsd.org Received: from boris.netgate.net (boris2.netgate.net [204.145.147.155]) by builder.freebsd.org (Postfix) with ESMTP id 3639541B0; Tue, 8 Feb 2000 02:24:29 -0800 (PST) Received: from localhost (wellsian@localhost) by boris.netgate.net (8.9.3/8.9.3) with ESMTP id CAA20177; Tue, 8 Feb 2000 02:23:00 -0800 (PST) (envelope-from wellsian@caffeine.com) Date: Tue, 8 Feb 2000 02:23:00 -0800 (PST) From: wellsian X-Sender: wellsian@boris.netgate.net To: James Holtom Cc: freebsd-questions@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: (SMP) Celerons + ABIT BP6 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org We have three of those combos and they work great. Those were the first boxes I used that could buildworld in under 30min - and at less than $275 for the board with 2 processors. (don't know now) The usual overclocking cautions apply strictly as fbsd will "use" both cpu's heavily. Check current or stable archives for more info (not bp6 specific). Basically, on an otherwise idle SMP system fbsd doesn't loop HLTs like it does without SMP. This exaggerates existing cooling problems for out of spec systems, exposing problems you might not otherwise notice until running gimps or something. I kind of like it. You know what you're working with. -Dave On Tue, 8 Feb 2000, James Holtom wrote: > Does anyone have any current [preferably good] experience using FreeBSD > 3.4 [Stable] on a twin-celeron ABIT BP6 board? > > I seem to recall seeing bad reports about stability and so on, but with > the mailing list archives dead ATM... > > Any info gratefully received... > > Regards, > > James To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 8 2:35:12 2000 Delivered-To: freebsd-smp@freebsd.org Received: from ms.tokyo.jcom.ne.jp (ms.tokyo.jcom.ne.jp [210.234.123.18]) by builder.freebsd.org (Postfix) with ESMTP id B1AFB41C4; Tue, 8 Feb 2000 02:34:53 -0800 (PST) Received: from daemon.local.idaemons.org (pc062023.tokyo.jcom.ne.jp [210.155.62.23]) by ms.tokyo.jcom.ne.jp (8.9.1/3.7W 06/01/99) with ESMTP id TAA16408; Tue, 8 Feb 2000 19:35:30 +0900 (JST) Received: by daemon.local.idaemons.org (8.9.3/3.7W) id TAA67341; Tue, 8 Feb 2000 19:34:59 +0900 (JST) Date: Tue, 08 Feb 2000 19:35:08 +0900 Message-ID: <86ln4wf0df.wl@archon.local.idaemons.org> From: "Akinori -Aki- MUSHA" To: jholtom@netforce.net Cc: freebsd-questions@freebsd.org, freebsd-smp@freebsd.org Subject: Re: (SMP) Celerons + ABIT BP6 In-Reply-To: In your message of "Tue, 8 Feb 2000 10:08:19 +0000 (GMT)" References: User-Agent: Wanderlust/2.2.17 (One Of Us) EMIKO/1.13.11 (Euglena viridis) FLIM/1.13.2 (Kasanui) APEL/10.1 MULE XEmacs/21.1 (patch 8) (Bryce Canyon) (i386--freebsd) Organization: Associated I. Daemons MIME-Version: 1.0 (generated by EMIKO 1.13.11 - "Euglena viridis") Content-Type: text/plain; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, At Tue, 8 Feb 2000 10:08:19 +0000 (GMT), James Holtom wrote: > Does anyone have any current [preferably good] experience using FreeBSD > 3.4 [Stable] on a twin-celeron ABIT BP6 board? I've ever heard no bad reports but so many good ones. The only thing you must take care will be the temperature. Do not overclock *too* much. :-) > I seem to recall seeing bad reports about stability and so on, but with > the mailing list archives dead ATM... FYI, Geocrawler is providing a great mailing lists archive. http://www.geocrawler.com/ http://www.geocrawler.com/lists/3/FreeBSD/ -- / /__ __ / ) ) ) ) / Akinori -Aki- MUSHA aka / (_ / ( (__( "If you choose not to decide you still have made a choice." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 8 7:51:31 2000 Delivered-To: freebsd-smp@freebsd.org Received: from disavowed.broken.net (disavowed.broken.net [204.216.142.33]) by builder.freebsd.org (Postfix) with ESMTP id 5A7764203; Tue, 8 Feb 2000 07:51:26 -0800 (PST) Received: (from ian@localhost) by disavowed.broken.net (8.8.7/8.8.7) id HAA06911; Tue, 8 Feb 2000 07:52:12 -0800 (PST) Date: Tue, 8 Feb 2000 07:52:11 -0800 (PST) From: Ian Struble To: James Holtom Cc: freebsd-questions@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: (SMP) Celerons + ABIT BP6 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I have had no problems with a 3.3-STABLE box with the configuration you describe (abit bp6, 256MB, dual 366@460(? can remember) cel). I don't think that a 3.4 setup would be any different. Ian On Tue, 8 Feb 2000, James Holtom wrote: > Does anyone have any current [preferably good] experience using FreeBSD > 3.4 [Stable] on a twin-celeron ABIT BP6 board? > > I seem to recall seeing bad reports about stability and so on, but with > the mailing list archives dead ATM... > > Any info gratefully received... > > Regards, > > James > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-questions" 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 Tue Feb 8 8:39:11 2000 Delivered-To: freebsd-smp@freebsd.org Received: from hotmail.com (f33.law8.hotmail.com [216.33.241.33]) by builder.freebsd.org (Postfix) with SMTP id F30383F65 for ; Tue, 8 Feb 2000 08:39:05 -0800 (PST) Received: (qmail 30405 invoked by uid 0); 8 Feb 2000 16:39:59 -0000 Message-ID: <20000208163959.30404.qmail@hotmail.com> Received: from 216.160.92.99 by www.hotmail.com with HTTP; Tue, 08 Feb 2000 08:39:59 PST X-Originating-IP: [216.160.92.99] Reply-To: the_hermit665@hotmail.com From: "Cosmic 665" To: jholtom@netforce.net, freebsd-questions@freebsd.org, freebsd-smp@freebsd.org Subject: Re: (SMP) Celerons + ABIT BP6 Date: Tue, 08 Feb 2000 08:39:59 PST Mime-Version: 1.0 Content-Type: text/plain; format=flowed Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org goto the FreeBSD SMP mailing list. I have got tons of threads on nothing but good reports on the Bp6. Matter in fact, I was gonna get one. But I already had PII's for the ASUS P2BD I got instead. :) >From: James Holtom >Reply-To: James Holtom >To: freebsd-questions@freebsd.org, freebsd-smp@freebsd.org >Subject: (SMP) Celerons + ABIT BP6 >Date: Tue, 8 Feb 2000 10:08:19 +0000 (GMT) >MIME-Version: 1.0 >Received: from [204.216.27.24] by hotmail.com (3.2) with ESMTP id >MHotMailBA693533001FD820F3BBCCD81B1813570; Tue Feb 08 02:04:35 2000 >Received: by builder.freebsd.org (Postfix, from userid 538)id 1161041A1; >Tue, 8 Feb 2000 02:01:46 -0800 (PST) >Received: from localhost (localhost [127.0.0.1])by builder.freebsd.org >(Postfix) with SMTPid AB975C9; Tue, 8 Feb 2000 02:01:46 -0800 (PST) >Received: by builder.freebsd.org (bulk_mailer v1.12); Tue, 8 Feb 2000 >02:01:46 -0800 >Received: from berk.mail.netforce.net (berk.mail.netforce.net >[195.58.64.236])by builder.freebsd.org (Postfix) with ESMTPid 06358419C; >Tue, 8 Feb 2000 02:01:40 -0800 (PST) >Received: from bagpuss.i.netforce.net (tarbuck.netforce.net >[195.58.64.34])by berk.mail.netforce.net (8.8.8/8.8.7) with ESMTP id >KAA12083;Tue, 8 Feb 2000 10:36:35 GMT >From owner-freebsd-smp@FreeBSD.ORG Tue Feb 08 02:06:48 2000 >Delivered-To: freebsd-smp@freebsd.org >Message-ID: > >Sender: owner-freebsd-smp@FreeBSD.ORG >X-Loop: FreeBSD.org >Precedence: bulk > >Does anyone have any current [preferably good] experience using FreeBSD >3.4 [Stable] on a twin-celeron ABIT BP6 board? > >I seem to recall seeing bad reports about stability and so on, but with >the mailing list archives dead ATM... > >Any info gratefully received... > >Regards, > >James > > > >To Unsubscribe: send mail to majordomo@FreeBSD.org >with "unsubscribe freebsd-smp" in the body of the message ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 8 12:11:47 2000 Delivered-To: freebsd-smp@freebsd.org Received: from internal.mail.demon.net (internal.mail.demon.net [193.195.224.3]) by builder.freebsd.org (Postfix) with ESMTP id E6A0040F5; Tue, 8 Feb 2000 12:11:22 -0800 (PST) Received: from gti.noc.demon.net (gti.noc.demon.net [195.11.55.101]) by internal.mail.demon.net with ESMTP id UAA03865; Tue, 8 Feb 2000 20:12:02 GMT Received: from localhost (kevinw@localhost) by gti.noc.demon.net (8.8.8/8.8.8) with ESMTP id UAA04293; Tue, 8 Feb 2000 20:11:59 GMT X-Authentication-Warning: gti.noc.demon.net: kevinw owned process doing -bs Date: Tue, 8 Feb 2000 20:11:59 +0000 (GMT) From: Kevin Walton X-Sender: kevinw@gti.noc.demon.net To: freebsd-smp@freebsd.org, freebsd-questions@freebsd.org Subject: 8 way Xeon SMP on a IBM NetFinity 8500 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi We are getting an 8 way Xeon IBM NetFinity 8500 and want to test FreeBSD on it. Just wondering: 1) Does FreeBSD support 8 way Xeon SMP? 2) How 'well' does it support it? 3) Has anyone tried it on a NetFinity 8500, and did they have any problems? Cheers Kevin -- Kevin Walton Team Leader of Interactive Services Thus Data Services Demon Internet http://www.games.demon.net/ http://www.demon.net/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Feb 9 0:22:37 2000 Delivered-To: freebsd-smp@freebsd.org Received: from prism.flugsvamp.com (cb58709-a.mdsn1.wi.home.com [24.17.241.9]) by builder.freebsd.org (Postfix) with ESMTP id 1551445B6 for ; Tue, 8 Feb 2000 18:02:53 -0800 (PST) Received: (from jlemon@localhost) by prism.flugsvamp.com (8.9.3/8.9.3) id UAA16653 for smp@freebsd.org; Tue, 8 Feb 2000 20:04:04 -0600 (CST) (envelope-from jlemon) Date: Tue, 8 Feb 2000 20:04:04 -0600 From: Jonathan Lemon To: smp@freebsd.org Subject: SMP on Dell Message-ID: <20000208200404.A16605@prism.flugsvamp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0pre2i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm trying to get a Dell Poweredge up and running SMP and am having some difficulty. Attached is the mptable from a uni-processor boot (which works), and the dmesg output from a failed SMP boot. As I'm currently lacking enough SMP clues to figure this out, could someone point me in the right direction? The 'Freeing (not implemented)' lines look suspicious to me. -- Jonathan =============================================================================== MPTable, version 2.0.15 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000fe710 signature: '_MP_' length: 16 bytes version: 1.4 checksum: 0x91 mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x000f0000 signature: 'PCMP' base table length: 652 version: 1.4 checksum: 0x03 OEM ID: 'DELL ' Product ID: 'POWEREDGE A2' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 70 local APIC address: 0xfee00000 extended table length: 208 extended table checksum: 241 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 3 0x11 BSP, usable 6 7 3 0x383fbff 0 0x11 AP, usable 6 7 3 0x383fbff 2 0x11 AP, usable 6 7 3 0x383fbff 1 0x11 AP, usable 6 7 3 0x383fbff -- Bus: Bus ID Type 0 PCI 1 PCI 2 PCI 3 PCI 4 PCI 5 PCI 6 PCI 7 PCI 8 PCI 9 PCI 10 PCI 11 PCI 12 PCI 13 PCI 14 PCI 15 PCI 16 PCI 17 ISA -- I/O APICs: APIC ID Version State Address 4 0x11 usable 0xfec00000 5 0x11 usable 0xfec01000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 17 0 4 0 INT conforms conforms 17 1 4 1 INT conforms conforms 17 3 4 3 INT conforms conforms 17 4 4 4 INT conforms conforms 17 6 4 6 INT conforms conforms 17 7 4 7 INT conforms conforms 17 2:A 4 8 INT conforms conforms 17 9 4 9 INT conforms conforms 17 12 4 12 INT conforms conforms 17 14 4 14 INT conforms conforms 17 15 4 15 INT conforms conforms 0 6:A 5 0 INT conforms conforms 0 5:A 5 1 INT conforms conforms 0 5:B 5 2 INT conforms conforms 12 12:A 5 3 INT conforms conforms 12 13:D 5 3 INT conforms conforms 12 12:D 5 4 INT conforms conforms 12 13:A 5 4 INT conforms conforms 0 7:A 5 5 INT conforms conforms 3 9:D 5 5 INT conforms conforms 3 10:C 5 5 INT conforms conforms 3 11:B 5 5 INT conforms conforms 3 8:A 5 6 INT conforms conforms 3 10:D 5 6 INT conforms conforms 3 11:C 5 6 INT conforms conforms 3 9:A 5 7 INT conforms conforms 3 11:D 5 7 INT conforms conforms 3 10:A 5 8 INT conforms conforms 0 7:B 5 9 INT conforms conforms 3 11:A 5 9 INT conforms conforms 0 8:A 5 10 INT conforms conforms 0 7:C 5 10 INT conforms conforms 3 8:B 5 10 INT conforms conforms 0 7:D 5 11 INT conforms conforms 3 8:C 5 11 INT conforms conforms 3 9:B 5 11 INT conforms conforms 3 8:D 5 12 INT conforms conforms 3 9:C 5 12 INT conforms conforms 3 10:B 5 12 INT conforms conforms 12 12:B 5 13 INT conforms conforms 12 13:C 5 13 INT conforms conforms 12 12:C 5 14 INT conforms conforms 12 13:B 5 14 INT conforms conforms 0 15:A 5 15 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 17 0 255 0 NMI active-hi edge 17 0 255 1 ------------------------------------------------------------------------------- MP Config Extended Table Entries: -- bus ID: 0 address type: I/O address address base: 0xd000 address range: 0x2000 -- bus ID: 0 address type: memory address address base: 0xa0000 address range: 0x20000 -- bus ID: 0 address type: I/O address address base: 0x0 address range: 0x6000 -- bus ID: 0 address type: memory address address base: 0xfb000000 address range: 0x1000000 -- bus ID: 0 address type: memory address address base: 0xfcd00000 address range: 0x1300000 -- bus ID: 0 address type: memory address address base: 0xfe000000 address range: 0x2000000 -- bus ID: 3 address type: I/O address address base: 0x9000 address range: 0x4000 -- bus ID: 3 address type: memory address address base: 0xf9e00000 address range: 0x1200000 -- bus ID: 12 address type: I/O address address base: 0x6000 address range: 0x3000 -- bus ID: 12 address type: memory address address base: 0xf8c00000 address range: 0x1200000 -- bus ID: 17 bus info: 0x01 parent bus ID: 0 ------------------------------------------------------------------------------- # SMP kernel config file options: # Required: options SMP # Symmetric MultiProcessor Kernel options APIC_IO # Symmetric (APIC) I/O # Optional (built-in defaults will work in most cases): #options NCPU=13 # number of CPUs #options NBUS=18 # number of busses #options NAPIC=2 # number of IO APICs #options NINTR=50 # number of INTs =============================================================================== ok boot -v SMAP type=01 base=00000000 00000000 len=00000000 000a0000 SMAP type=01 base=00000000 00100000 len=00000000 f7efe000 SMAP type=02 base=00000000 f7ffe000 len=00000000 00002000 SMAP type=02 base=00000000 fec00000 len=00000000 00010000 SMAP type=02 base=00000000 fee00000 len=00000000 00010000 SMAP type=02 base=00000000 fff80000 len=00000000 00080000 Copyright (c) 1992-2000 The FreeBSD Project. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 4.0-CURRENT #8: Sun Feb 6 14:16:45 PST 2000 jlemon@XXXXXXXXXXXXXX:/usr/src/sys/compile/SMP4 Calibrating clock(s) ... TSC clock: 549468250 Hz, i8254 clock: 1193244 Hz CLK_USE_I8254_CALIBRATION not specified - using default frequency Timecounter "i8254" frequency 1193182 Hz CLK_USE_TSC_CALIBRATION not specified - using old calibration method CPU: Pentium III/Xeon (549.44-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x673 Stepping = 3 Features=0x383fbff real memory = 4160741376 (4063224K bytes) Physical memory chunk(s): 0x00001000 - 0x0009efff, 647168 bytes (158 pages) 0x00352000 - 0xf7ffbfff, 4157251584 bytes (4294933674 pages) avail memory = 4038045696 (3943404K bytes) Programming 16 pins in IOAPIC #0 Programming 16 pins in IOAPIC #1 IOAPIC #1 intpin 0 -> irq 2 IOAPIC #1 intpin 1 -> irq 5 IOAPIC #1 intpin 2 -> irq 10 IOAPIC #1 intpin 3 -> irq 11 IOAPIC #1 intpin 4 -> irq 13 IOAPIC #1 intpin 5 -> irq 16 IOAPIC #1 intpin 6 -> irq 17 IOAPIC #1 intpin 7 -> irq 18 IOAPIC #1 intpin 8 -> irq 19 IOAPIC #1 intpin 9 -> irq 20 IOAPIC #1 intpin 10 -> irq 21 IOAPIC #1 intpin 11 -> irq 22 IOAPIC #1 intpin 12 -> irq 23 SMP: CPU0 apic_initialize(): lint0: 0x00000700 lint1: 0x00010400 TPR: 0x00000010 SVR: 0x000001ff FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 3, version: 0x00040011, at 0xfee00000 cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 cpu2 (AP): apic id: 1, version: 0x00040011, at 0xfee00000 cpu3 (AP): apic id: 2, version: 0x00040011, at 0xfee00000 io0 (APIC): apic id: 4, version: 0x000f0011, at 0xfec00000 io1 (APIC): apic id: 5, version: 0x000f0011, at 0xfec01000 bios32: Found BIOS32 Service Directory header at 0xc00ffe80 bios32: Entry = 0xffe90 (c00ffe90) Rev = 0 Len = 1 pcibios: PCI BIOS entry at 0xc7fe pnpbios: Found PnP BIOS data at 0xc00fe2d0 pnpbios: Entry = f0000:e2f4 Rev = 1.0 Other BIOS signatures found: ACPI: 000fdcc0 Preloaded elf kernel "kernel.DELL-SMP" at 0xc0336000. Pentium Pro MTRR support enabled md0: Malloc disk Creating DISK md0 Math emulator present SMP: CPU0 bsp_apic_configure(): lint0: 0x00010700 lint1: 0x00000400 TPR: 0x00000010 SVR: 0x000001ff pci_open(1): mode 1 addr port (0x0cf8) is 0x80000074 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) devclass_alloc_unit: pcib3 already exists, using next available unit number npx0: on motherboard npx0: INT 16 interface pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib3: on motherboard Freeing (NOT implemented) redirected PCI irq 11. found-> vendor=0x1385, dev=0x620a, revid=0x01 class=02-00-00, hdrtype=0x00, mfdev=0 subordinatebus=0 secondarybus=0 intpin=a, irq=20 map[10]: type 1, range 32, base faefc000, size 14 pci3: on pcib3 ti0: irq 20 at device 11.0 on pci3 ti0: Ethernet address: 00:a0:cc:73:35:88 bpf: ti0 attached pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib0: on motherboard found-> vendor=0x1166, dev=0x0008, revid=0x21 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0008, revid=0x01 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0006, revid=0x00 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0006, revid=0x00 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1002, dev=0x4759, revid=0x7a class=03-00-00, hdrtype=0x00, mfdev=0 subordinatebus=0 secondarybus=0 map[10]: type 1, range 32, base fb000000, size 24 map[14]: type 1, range 32, base 0000ec00, size 8 map[18]: type 1, range 32, base fdeff000, size 12 Freeing (NOT implemented) redirected PCI irq 11. found-> vendor=0x9005, dev=0x00cf, revid=0x01 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=a, irq=5 map[10]: type 1, range 32, base 0000e800, size 8 map[14]: type 1, range 64, base fdefe000, size 12 found-> vendor=0x9005, dev=0x00cf, revid=0x01 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=b, irq=10 map[10]: type 1, range 32, base 0000e400, size 8 map[14]: type 1, range 64, base fdefd000, size 12 Freeing (NOT implemented) redirected PCI irq 5. found-> vendor=0x8086, dev=0x1229, revid=0x08 class=02-00-00, hdrtype=0x00, mfdev=0 subordinatebus=0 secondarybus=0 intpin=a, irq=21 map[10]: type 1, range 32, base fdefc000, size 12 map[14]: type 1, range 32, base 0000e0c0, size 6 map[18]: type 1, range 32, base fdd00000, size 20 found-> vendor=0x1166, dev=0x0200, revid=0x4f class=06-01-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0211, revid=0x00 class=01-01-8a, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 map[20]: type 1, range 32, base 000008b0, size 4 pci0: on pcib0 vga-pci0: at device 4.0 on pci0 ahc0: irq 5 at device 5.0 on pci0 OptionMode = 3 ahc0: Reading SEEPROM...done. ahc0: Manual LVD Termination ahc0: BIOS eeprom is present ahc0: Secondary High byte termination Enabled ahc0: Secondary Low byte termination Enabled ahc0: Primary Low Byte termination Enabled ahc0: Primary High Byte termination Enabled ahc0: aic7899 Wide Channel A, SCSI Id=7, 16/255 SCBs ahc0: Downloading Sequencer Program... 399 instructions downloaded ahc1: irq 10 at device 5.1 on pci0 OptionMode = 3 ahc1: Reading SEEPROM...done. ahc1: Manual LVD Termination ahc1: BIOS eeprom is present ahc1: Secondary High byte termination Enabled ahc1: Secondary Low byte termination Enabled ahc1: Primary Low Byte termination Enabled ahc1: Primary High Byte termination Enabled ahc1: aic7899 Wide Channel B, SCSI Id=7, 16/255 SCBs ahc1: Downloading Sequencer Program... 399 instructions downloaded fxp0: irq 21 at device 8.0 on pci0 fxp0: Ethernet address 00:c0:4f:a0:46:1a bpf: fxp0 attached isab0: at device 15.0 on pci0 isa0: on isab0 pci0: Unknown PCI ATA controller (vendor=0x1166, dev=0x0211) at 15.1 pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib4: on motherboard pci4: on pcib4 pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib12: on motherboard Freeing (NOT implemented) redirected PCI irq 10. found-> vendor=0x1000, dev=0x000b, revid=0x05 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=a, irq=11 map[10]: type 1, range 32, base 00008c00, size 8 map[14]: type 1, range 64, base f9cffc00, size 10 map[1c]: type 1, range 64, base f9cfc000, size 13 Freeing (NOT implemented) redirected PCI irq 5. found-> vendor=0x1000, dev=0x000b, revid=0x05 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=b, irq=255 map[10]: type 1, range 32, base 00008800, size 8 map[14]: type 1, range 64, base f9cff800, size 10 map[1c]: type 1, range 64, base f9cfa000, size 13 pci12: on pcib12 sym0: <896> irq 11 at device 12.0 on pci12 sym0: Symbios NVRAM, ID 7, Fast-40, LVD, parity checking sym0: open drain IRQ line driver, using on-chip SRAM sym0: handling phase mismatch from SCRIPTS. sym0: initial SCNTL3/DMODE/DCNTL/CTEST3/4/5 = (hex) 07/4e/a0/01/00/24 sym0: final SCNTL3/DMODE/DCNTL/CTEST3/4/5 = (hex) 07/4e/80/01/08/24 sym0: Delay (GEN=11): 243 msec, 36577 KHz sym0: Delay (GEN=11): 276 msec, 32204 KHz sym0: Delay (GEN=11): 276 msec, 32204 KHz sym0: enabling clock multiplier sym0: Downloading SCSI SCRIPTS. sym1: <896> at device 12.1 on pci12 sym1: failed to allocate IRQ resource Fatal trap 12: page fault while in kernel mode mp_lock = 00000002; cpuid = 0; lapic.id = 03000000 fault virtual address = 0x0 fault code = supervisor read, page not present instruction pointer = 0x8:0xc01f5091 stack pointer = 0x10:0xc034ace8 frame pointer = 0x10:0xc034acf8 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 0 (swapper) interrupt mask = net tty bio cam <- SMP: XXX trap number = 12 panic: page fault mp_lock = 00000002; cpuid = 0; lapic.id = 03000000 Uptime: 0s Automatic reboot in 15 seconds - press a key on the console to abort Rebooting... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Feb 9 8:31:57 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mass.cdrom.com (castles560.castles.com [208.214.165.124]) by builder.freebsd.org (Postfix) with ESMTP id 0928E4286 for ; Wed, 9 Feb 2000 08:31:36 -0800 (PST) Received: from mass.cdrom.com (localhost [127.0.0.1]) by mass.cdrom.com (8.9.3/8.9.3) with ESMTP id IAA00710; Wed, 9 Feb 2000 08:41:48 -0800 (PST) (envelope-from msmith@mass.cdrom.com) Message-Id: <200002091641.IAA00710@mass.cdrom.com> X-Mailer: exmh version 2.1.1 10/15/1999 To: Kevin Walton Cc: freebsd-smp@freebsd.org Subject: Re: 8 way Xeon SMP on a IBM NetFinity 8500 In-reply-to: Your message of "Tue, 08 Feb 2000 20:11:59 GMT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 09 Feb 2000 08:41:47 -0800 From: Mike Smith Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Hi > > We are getting an 8 way Xeon IBM NetFinity 8500 and want to test FreeBSD > on it. Just wondering: > > 1) Does FreeBSD support 8 way Xeon SMP? > 2) How 'well' does it support it? > 3) Has anyone tried it on a NetFinity 8500, and did they have any > problems? I expect that this will be a Corollary-based system, and probably not Intel MP-spec compliant. If that's the case, we won't on it at all. -- \\ Give a man a fish, and you feed him for a day. \\ Mike Smith \\ Tell him he should learn how to fish himself, \\ msmith@freebsd.org \\ and he'll hate you for a lifetime. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Feb 9 10:14:17 2000 Delivered-To: freebsd-smp@freebsd.org Received: from prism.flugsvamp.com (cb58709-a.mdsn1.wi.home.com [24.17.241.9]) by builder.freebsd.org (Postfix) with ESMTP id F1D9C40A3 for ; Wed, 9 Feb 2000 10:14:05 -0800 (PST) Received: (from jlemon@localhost) by prism.flugsvamp.com (8.9.3/8.9.3) id LAA17665 for smp@freebsd.org; Wed, 9 Feb 2000 11:45:06 -0600 (CST) (envelope-from jlemon) Date: Wed, 9 Feb 2000 11:45:06 -0600 From: Jonathan Lemon To: smp@freebsd.org Subject: SMP on Dell Message-ID: <20000209114506.D17611@prism.flugsvamp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0pre2i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm trying to get a Dell Poweredge up and running SMP and am having some difficulty. Attached is the mptable from a uni-processor boot (which works), and the dmesg output from a failed SMP boot. As I'm currently lacking enough SMP clues to figure this out, could someone point me in the right direction? The 'Freeing (not implemented)' lines look suspicious to me. -- Jonathan =============================================================================== MPTable, version 2.0.15 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000fe710 signature: '_MP_' length: 16 bytes version: 1.4 checksum: 0x91 mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x000f0000 signature: 'PCMP' base table length: 652 version: 1.4 checksum: 0x03 OEM ID: 'DELL ' Product ID: 'POWEREDGE A2' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 70 local APIC address: 0xfee00000 extended table length: 208 extended table checksum: 241 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 3 0x11 BSP, usable 6 7 3 0x383fbff 0 0x11 AP, usable 6 7 3 0x383fbff 2 0x11 AP, usable 6 7 3 0x383fbff 1 0x11 AP, usable 6 7 3 0x383fbff -- Bus: Bus ID Type 0 PCI 1 PCI 2 PCI 3 PCI 4 PCI 5 PCI 6 PCI 7 PCI 8 PCI 9 PCI 10 PCI 11 PCI 12 PCI 13 PCI 14 PCI 15 PCI 16 PCI 17 ISA -- I/O APICs: APIC ID Version State Address 4 0x11 usable 0xfec00000 5 0x11 usable 0xfec01000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 17 0 4 0 INT conforms conforms 17 1 4 1 INT conforms conforms 17 3 4 3 INT conforms conforms 17 4 4 4 INT conforms conforms 17 6 4 6 INT conforms conforms 17 7 4 7 INT conforms conforms 17 2:A 4 8 INT conforms conforms 17 9 4 9 INT conforms conforms 17 12 4 12 INT conforms conforms 17 14 4 14 INT conforms conforms 17 15 4 15 INT conforms conforms 0 6:A 5 0 INT conforms conforms 0 5:A 5 1 INT conforms conforms 0 5:B 5 2 INT conforms conforms 12 12:A 5 3 INT conforms conforms 12 13:D 5 3 INT conforms conforms 12 12:D 5 4 INT conforms conforms 12 13:A 5 4 INT conforms conforms 0 7:A 5 5 INT conforms conforms 3 9:D 5 5 INT conforms conforms 3 10:C 5 5 INT conforms conforms 3 11:B 5 5 INT conforms conforms 3 8:A 5 6 INT conforms conforms 3 10:D 5 6 INT conforms conforms 3 11:C 5 6 INT conforms conforms 3 9:A 5 7 INT conforms conforms 3 11:D 5 7 INT conforms conforms 3 10:A 5 8 INT conforms conforms 0 7:B 5 9 INT conforms conforms 3 11:A 5 9 INT conforms conforms 0 8:A 5 10 INT conforms conforms 0 7:C 5 10 INT conforms conforms 3 8:B 5 10 INT conforms conforms 0 7:D 5 11 INT conforms conforms 3 8:C 5 11 INT conforms conforms 3 9:B 5 11 INT conforms conforms 3 8:D 5 12 INT conforms conforms 3 9:C 5 12 INT conforms conforms 3 10:B 5 12 INT conforms conforms 12 12:B 5 13 INT conforms conforms 12 13:C 5 13 INT conforms conforms 12 12:C 5 14 INT conforms conforms 12 13:B 5 14 INT conforms conforms 0 15:A 5 15 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 17 0 255 0 NMI active-hi edge 17 0 255 1 ------------------------------------------------------------------------------- ok boot -v SMAP type=01 base=00000000 00000000 len=00000000 000a0000 SMAP type=01 base=00000000 00100000 len=00000000 f7efe000 SMAP type=02 base=00000000 f7ffe000 len=00000000 00002000 SMAP type=02 base=00000000 fec00000 len=00000000 00010000 SMAP type=02 base=00000000 fee00000 len=00000000 00010000 SMAP type=02 base=00000000 fff80000 len=00000000 00080000 Copyright (c) 1992-2000 The FreeBSD Project. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 4.0-CURRENT #8: Sun Feb 6 14:16:45 PST 2000 jlemon@XXXXXXXXXXXXXX:/usr/src/sys/compile/SMP4 Calibrating clock(s) ... TSC clock: 549468250 Hz, i8254 clock: 1193244 Hz CLK_USE_I8254_CALIBRATION not specified - using default frequency Timecounter "i8254" frequency 1193182 Hz CLK_USE_TSC_CALIBRATION not specified - using old calibration method CPU: Pentium III/Xeon (549.44-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x673 Stepping = 3 Features=0x383fbff real memory = 4160741376 (4063224K bytes) Physical memory chunk(s): 0x00001000 - 0x0009efff, 647168 bytes (158 pages) 0x00352000 - 0xf7ffbfff, 4157251584 bytes (4294933674 pages) avail memory = 4038045696 (3943404K bytes) Programming 16 pins in IOAPIC #0 Programming 16 pins in IOAPIC #1 IOAPIC #1 intpin 0 -> irq 2 IOAPIC #1 intpin 1 -> irq 5 IOAPIC #1 intpin 2 -> irq 10 IOAPIC #1 intpin 3 -> irq 11 IOAPIC #1 intpin 4 -> irq 13 IOAPIC #1 intpin 5 -> irq 16 IOAPIC #1 intpin 6 -> irq 17 IOAPIC #1 intpin 7 -> irq 18 IOAPIC #1 intpin 8 -> irq 19 IOAPIC #1 intpin 9 -> irq 20 IOAPIC #1 intpin 10 -> irq 21 IOAPIC #1 intpin 11 -> irq 22 IOAPIC #1 intpin 12 -> irq 23 SMP: CPU0 apic_initialize(): lint0: 0x00000700 lint1: 0x00010400 TPR: 0x00000010 SVR: 0x000001ff FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 3, version: 0x00040011, at 0xfee00000 cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 cpu2 (AP): apic id: 1, version: 0x00040011, at 0xfee00000 cpu3 (AP): apic id: 2, version: 0x00040011, at 0xfee00000 io0 (APIC): apic id: 4, version: 0x000f0011, at 0xfec00000 io1 (APIC): apic id: 5, version: 0x000f0011, at 0xfec01000 bios32: Found BIOS32 Service Directory header at 0xc00ffe80 bios32: Entry = 0xffe90 (c00ffe90) Rev = 0 Len = 1 pcibios: PCI BIOS entry at 0xc7fe pnpbios: Found PnP BIOS data at 0xc00fe2d0 pnpbios: Entry = f0000:e2f4 Rev = 1.0 Other BIOS signatures found: ACPI: 000fdcc0 Preloaded elf kernel "kernel.DELL-SMP" at 0xc0336000. Pentium Pro MTRR support enabled md0: Malloc disk Creating DISK md0 Math emulator present SMP: CPU0 bsp_apic_configure(): lint0: 0x00010700 lint1: 0x00000400 TPR: 0x00000010 SVR: 0x000001ff pci_open(1): mode 1 addr port (0x0cf8) is 0x80000074 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) devclass_alloc_unit: pcib3 already exists, using next available unit number npx0: on motherboard npx0: INT 16 interface pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib3: on motherboard Freeing (NOT implemented) redirected PCI irq 11. found-> vendor=0x1385, dev=0x620a, revid=0x01 class=02-00-00, hdrtype=0x00, mfdev=0 subordinatebus=0 secondarybus=0 intpin=a, irq=20 map[10]: type 1, range 32, base faefc000, size 14 pci3: on pcib3 ti0: irq 20 at device 11.0 on pci3 ti0: Ethernet address: 00:a0:cc:73:35:88 bpf: ti0 attached pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib0: on motherboard found-> vendor=0x1166, dev=0x0008, revid=0x21 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0008, revid=0x01 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0006, revid=0x00 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0006, revid=0x00 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1002, dev=0x4759, revid=0x7a class=03-00-00, hdrtype=0x00, mfdev=0 subordinatebus=0 secondarybus=0 map[10]: type 1, range 32, base fb000000, size 24 map[14]: type 1, range 32, base 0000ec00, size 8 map[18]: type 1, range 32, base fdeff000, size 12 Freeing (NOT implemented) redirected PCI irq 11. found-> vendor=0x9005, dev=0x00cf, revid=0x01 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=a, irq=5 map[10]: type 1, range 32, base 0000e800, size 8 map[14]: type 1, range 64, base fdefe000, size 12 found-> vendor=0x9005, dev=0x00cf, revid=0x01 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=b, irq=10 map[10]: type 1, range 32, base 0000e400, size 8 map[14]: type 1, range 64, base fdefd000, size 12 Freeing (NOT implemented) redirected PCI irq 5. found-> vendor=0x8086, dev=0x1229, revid=0x08 class=02-00-00, hdrtype=0x00, mfdev=0 subordinatebus=0 secondarybus=0 intpin=a, irq=21 map[10]: type 1, range 32, base fdefc000, size 12 map[14]: type 1, range 32, base 0000e0c0, size 6 map[18]: type 1, range 32, base fdd00000, size 20 found-> vendor=0x1166, dev=0x0200, revid=0x4f class=06-01-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0211, revid=0x00 class=01-01-8a, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 map[20]: type 1, range 32, base 000008b0, size 4 pci0: on pcib0 vga-pci0: at device 4.0 on pci0 ahc0: irq 5 at device 5.0 on pci0 OptionMode = 3 ahc0: Reading SEEPROM...done. ahc0: Manual LVD Termination ahc0: BIOS eeprom is present ahc0: Secondary High byte termination Enabled ahc0: Secondary Low byte termination Enabled ahc0: Primary Low Byte termination Enabled ahc0: Primary High Byte termination Enabled ahc0: aic7899 Wide Channel A, SCSI Id=7, 16/255 SCBs ahc0: Downloading Sequencer Program... 399 instructions downloaded ahc1: irq 10 at device 5.1 on pci0 OptionMode = 3 ahc1: Reading SEEPROM...done. ahc1: Manual LVD Termination ahc1: BIOS eeprom is present ahc1: Secondary High byte termination Enabled ahc1: Secondary Low byte termination Enabled ahc1: Primary Low Byte termination Enabled ahc1: Primary High Byte termination Enabled ahc1: aic7899 Wide Channel B, SCSI Id=7, 16/255 SCBs ahc1: Downloading Sequencer Program... 399 instructions downloaded fxp0: irq 21 at device 8.0 on pci0 fxp0: Ethernet address 00:c0:4f:a0:46:1a bpf: fxp0 attached isab0: at device 15.0 on pci0 isa0: on isab0 pci0: Unknown PCI ATA controller (vendor=0x1166, dev=0x0211) at 15.1 pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib4: on motherboard pci4: on pcib4 pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib12: on motherboard Freeing (NOT implemented) redirected PCI irq 10. found-> vendor=0x1000, dev=0x000b, revid=0x05 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=a, irq=11 map[10]: type 1, range 32, base 00008c00, size 8 map[14]: type 1, range 64, base f9cffc00, size 10 map[1c]: type 1, range 64, base f9cfc000, size 13 Freeing (NOT implemented) redirected PCI irq 5. found-> vendor=0x1000, dev=0x000b, revid=0x05 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=b, irq=255 map[10]: type 1, range 32, base 00008800, size 8 map[14]: type 1, range 64, base f9cff800, size 10 map[1c]: type 1, range 64, base f9cfa000, size 13 pci12: on pcib12 sym0: <896> irq 11 at device 12.0 on pci12 sym0: Symbios NVRAM, ID 7, Fast-40, LVD, parity checking sym0: open drain IRQ line driver, using on-chip SRAM sym0: handling phase mismatch from SCRIPTS. sym0: initial SCNTL3/DMODE/DCNTL/CTEST3/4/5 = (hex) 07/4e/a0/01/00/24 sym0: final SCNTL3/DMODE/DCNTL/CTEST3/4/5 = (hex) 07/4e/80/01/08/24 sym0: Delay (GEN=11): 243 msec, 36577 KHz sym0: Delay (GEN=11): 276 msec, 32204 KHz sym0: Delay (GEN=11): 276 msec, 32204 KHz sym0: enabling clock multiplier sym0: Downloading SCSI SCRIPTS. sym1: <896> at device 12.1 on pci12 sym1: failed to allocate IRQ resource Fatal trap 12: page fault while in kernel mode mp_lock = 00000002; cpuid = 0; lapic.id = 03000000 fault virtual address = 0x0 fault code = supervisor read, page not present instruction pointer = 0x8:0xc01f5091 stack pointer = 0x10:0xc034ace8 frame pointer = 0x10:0xc034acf8 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 0 (swapper) interrupt mask = net tty bio cam <- SMP: XXX trap number = 12 panic: page fault mp_lock = 00000002; cpuid = 0; lapic.id = 03000000 Uptime: 0s Automatic reboot in 15 seconds - press a key on the console to abort Rebooting... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Feb 9 10:45:38 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mel.alcatel.fr (mel.alcatel.fr [212.208.74.132]) by builder.freebsd.org (Postfix) with ESMTP id 4B899413A for ; Wed, 9 Feb 2000 10:45:19 -0800 (PST) Received: from aifhs10.alcatel.fr (mailhub2.alcatel.fr [155.132.188.80]) by mel.alcatel.fr (ALCANET/SMTP) with ESMTP id TAA17943; Wed, 9 Feb 2000 19:37:19 +0100 From: Thierry.Herbelot@alcatel.fr Received: from frmta003.netfr.alcatel.fr (frmta003.netfr.alcatel.fr [155.132.251.32]) by aifhs10.alcatel.fr (ALCANET/SMTP2) with SMTP id TAA19579; Wed, 9 Feb 2000 19:39:21 +0100 (MET) Received: by frmta003.netfr.alcatel.fr(Lotus SMTP MTA v4.6.6 (890.1 7-16-1999)) id C1256880.0066F6C0 ; Wed, 9 Feb 2000 19:44:38 +0100 X-Lotus-FromDomain: ALCATEL To: Jonathan Lemon Cc: smp@FreeBSD.ORG Message-ID: Date: Wed, 9 Feb 2000 19:44:33 +0100 Subject: Re: SMP on Dell Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Content-Disposition: inline Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello, I'm quite suspicious seeing hints of a non-supported SMP chipset from Renaissance : pcib4: on motherboard pci4: on pcib4 pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib12: on motherboard If it is like I think, you may be out of luck (which is bad because yours is quite a fine beast) (but I may be dead wrong) TfH To: smp@FreeBSD.ORG cc: (bcc: Thierry HERBELOT/FR/ALCATEL) Subject: SMP on Dell I'm trying to get a Dell Poweredge up and running SMP and am having some difficulty. Attached is the mptable from a uni-processor boot (which works), and the dmesg output from a failed SMP boot. As I'm currently lacking enough SMP clues to figure this out, could someone point me in the right direction? The 'Freeing (not implemented)' lines look suspicious to me. -- Jonathan =============================================================================== MPTable, version 2.0.15 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000fe710 signature: '_MP_' length: 16 bytes version: 1.4 checksum: 0x91 mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x000f0000 signature: 'PCMP' base table length: 652 version: 1.4 checksum: 0x03 OEM ID: 'DELL ' Product ID: 'POWEREDGE A2' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 70 local APIC address: 0xfee00000 extended table length: 208 extended table checksum: 241 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 3 0x11 BSP, usable 6 7 3 0x383fbff 0 0x11 AP, usable 6 7 3 0x383fbff 2 0x11 AP, usable 6 7 3 0x383fbff 1 0x11 AP, usable 6 7 3 0x383fbff -- Bus: Bus ID Type 0 PCI 1 PCI 2 PCI 3 PCI 4 PCI 5 PCI 6 PCI 7 PCI 8 PCI 9 PCI 10 PCI 11 PCI 12 PCI 13 PCI 14 PCI 15 PCI 16 PCI 17 ISA -- I/O APICs: APIC ID Version State Address 4 0x11 usable 0xfec00000 5 0x11 usable 0xfec01000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 17 0 4 0 INT conforms conforms 17 1 4 1 INT conforms conforms 17 3 4 3 INT conforms conforms 17 4 4 4 INT conforms conforms 17 6 4 6 INT conforms conforms 17 7 4 7 INT conforms conforms 17 2:A 4 8 INT conforms conforms 17 9 4 9 INT conforms conforms 17 12 4 12 INT conforms conforms 17 14 4 14 INT conforms conforms 17 15 4 15 INT conforms conforms 0 6:A 5 0 INT conforms conforms 0 5:A 5 1 INT conforms conforms 0 5:B 5 2 INT conforms conforms 12 12:A 5 3 INT conforms conforms 12 13:D 5 3 INT conforms conforms 12 12:D 5 4 INT conforms conforms 12 13:A 5 4 INT conforms conforms 0 7:A 5 5 INT conforms conforms 3 9:D 5 5 INT conforms conforms 3 10:C 5 5 INT conforms conforms 3 11:B 5 5 INT conforms conforms 3 8:A 5 6 INT conforms conforms 3 10:D 5 6 INT conforms conforms 3 11:C 5 6 INT conforms conforms 3 9:A 5 7 INT conforms conforms 3 11:D 5 7 INT conforms conforms 3 10:A 5 8 INT conforms conforms 0 7:B 5 9 INT conforms conforms 3 11:A 5 9 INT conforms conforms 0 8:A 5 10 INT conforms conforms 0 7:C 5 10 INT conforms conforms 3 8:B 5 10 INT conforms conforms 0 7:D 5 11 INT conforms conforms 3 8:C 5 11 INT conforms conforms 3 9:B 5 11 INT conforms conforms 3 8:D 5 12 INT conforms conforms 3 9:C 5 12 INT conforms conforms 3 10:B 5 12 INT conforms conforms 12 12:B 5 13 INT conforms conforms 12 13:C 5 13 INT conforms conforms 12 12:C 5 14 INT conforms conforms 12 13:B 5 14 INT conforms conforms 0 15:A 5 15 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 17 0 255 0 NMI active-hi edge 17 0 255 1 ------------------------------------------------------------------------------- ok boot -v SMAP type=01 base=00000000 00000000 len=00000000 000a0000 SMAP type=01 base=00000000 00100000 len=00000000 f7efe000 SMAP type=02 base=00000000 f7ffe000 len=00000000 00002000 SMAP type=02 base=00000000 fec00000 len=00000000 00010000 SMAP type=02 base=00000000 fee00000 len=00000000 00010000 SMAP type=02 base=00000000 fff80000 len=00000000 00080000 Copyright (c) 1992-2000 The FreeBSD Project. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 4.0-CURRENT #8: Sun Feb 6 14:16:45 PST 2000 jlemon@XXXXXXXXXXXXXX:/usr/src/sys/compile/SMP4 Calibrating clock(s) ... TSC clock: 549468250 Hz, i8254 clock: 1193244 Hz CLK_USE_I8254_CALIBRATION not specified - using default frequency Timecounter "i8254" frequency 1193182 Hz CLK_USE_TSC_CALIBRATION not specified - using old calibration method CPU: Pentium III/Xeon (549.44-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x673 Stepping = 3 Features=0x383fbff real memory = 4160741376 (4063224K bytes) Physical memory chunk(s): 0x00001000 - 0x0009efff, 647168 bytes (158 pages) 0x00352000 - 0xf7ffbfff, 4157251584 bytes (4294933674 pages) avail memory = 4038045696 (3943404K bytes) Programming 16 pins in IOAPIC #0 Programming 16 pins in IOAPIC #1 IOAPIC #1 intpin 0 -> irq 2 IOAPIC #1 intpin 1 -> irq 5 IOAPIC #1 intpin 2 -> irq 10 IOAPIC #1 intpin 3 -> irq 11 IOAPIC #1 intpin 4 -> irq 13 IOAPIC #1 intpin 5 -> irq 16 IOAPIC #1 intpin 6 -> irq 17 IOAPIC #1 intpin 7 -> irq 18 IOAPIC #1 intpin 8 -> irq 19 IOAPIC #1 intpin 9 -> irq 20 IOAPIC #1 intpin 10 -> irq 21 IOAPIC #1 intpin 11 -> irq 22 IOAPIC #1 intpin 12 -> irq 23 SMP: CPU0 apic_initialize(): lint0: 0x00000700 lint1: 0x00010400 TPR: 0x00000010 SVR: 0x000001ff FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 3, version: 0x00040011, at 0xfee00000 cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 cpu2 (AP): apic id: 1, version: 0x00040011, at 0xfee00000 cpu3 (AP): apic id: 2, version: 0x00040011, at 0xfee00000 io0 (APIC): apic id: 4, version: 0x000f0011, at 0xfec00000 io1 (APIC): apic id: 5, version: 0x000f0011, at 0xfec01000 bios32: Found BIOS32 Service Directory header at 0xc00ffe80 bios32: Entry = 0xffe90 (c00ffe90) Rev = 0 Len = 1 pcibios: PCI BIOS entry at 0xc7fe pnpbios: Found PnP BIOS data at 0xc00fe2d0 pnpbios: Entry = f0000:e2f4 Rev = 1.0 Other BIOS signatures found: ACPI: 000fdcc0 Preloaded elf kernel "kernel.DELL-SMP" at 0xc0336000. Pentium Pro MTRR support enabled md0: Malloc disk Creating DISK md0 Math emulator present SMP: CPU0 bsp_apic_configure(): lint0: 0x00010700 lint1: 0x00000400 TPR: 0x00000010 SVR: 0x000001ff pci_open(1): mode 1 addr port (0x0cf8) is 0x80000074 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) devclass_alloc_unit: pcib3 already exists, using next available unit number npx0: on motherboard npx0: INT 16 interface pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib3: on motherboard Freeing (NOT implemented) redirected PCI irq 11. found-> vendor=0x1385, dev=0x620a, revid=0x01 class=02-00-00, hdrtype=0x00, mfdev=0 subordinatebus=0 secondarybus=0 intpin=a, irq=20 map[10]: type 1, range 32, base faefc000, size 14 pci3: on pcib3 ti0: irq 20 at device 11.0 on pci3 ti0: Ethernet address: 00:a0:cc:73:35:88 bpf: ti0 attached pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib0: on motherboard found-> vendor=0x1166, dev=0x0008, revid=0x21 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0008, revid=0x01 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0006, revid=0x00 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0006, revid=0x00 class=06-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1002, dev=0x4759, revid=0x7a class=03-00-00, hdrtype=0x00, mfdev=0 subordinatebus=0 secondarybus=0 map[10]: type 1, range 32, base fb000000, size 24 map[14]: type 1, range 32, base 0000ec00, size 8 map[18]: type 1, range 32, base fdeff000, size 12 Freeing (NOT implemented) redirected PCI irq 11. found-> vendor=0x9005, dev=0x00cf, revid=0x01 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=a, irq=5 map[10]: type 1, range 32, base 0000e800, size 8 map[14]: type 1, range 64, base fdefe000, size 12 found-> vendor=0x9005, dev=0x00cf, revid=0x01 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=b, irq=10 map[10]: type 1, range 32, base 0000e400, size 8 map[14]: type 1, range 64, base fdefd000, size 12 Freeing (NOT implemented) redirected PCI irq 5. found-> vendor=0x8086, dev=0x1229, revid=0x08 class=02-00-00, hdrtype=0x00, mfdev=0 subordinatebus=0 secondarybus=0 intpin=a, irq=21 map[10]: type 1, range 32, base fdefc000, size 12 map[14]: type 1, range 32, base 0000e0c0, size 6 map[18]: type 1, range 32, base fdd00000, size 20 found-> vendor=0x1166, dev=0x0200, revid=0x4f class=06-01-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 found-> vendor=0x1166, dev=0x0211, revid=0x00 class=01-01-8a, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 map[20]: type 1, range 32, base 000008b0, size 4 pci0: on pcib0 vga-pci0: at device 4.0 on pci0 ahc0: irq 5 at device 5.0 on pci0 OptionMode = 3 ahc0: Reading SEEPROM...done. ahc0: Manual LVD Termination ahc0: BIOS eeprom is present ahc0: Secondary High byte termination Enabled ahc0: Secondary Low byte termination Enabled ahc0: Primary Low Byte termination Enabled ahc0: Primary High Byte termination Enabled ahc0: aic7899 Wide Channel A, SCSI Id=7, 16/255 SCBs ahc0: Downloading Sequencer Program... 399 instructions downloaded ahc1: irq 10 at device 5.1 on pci0 OptionMode = 3 ahc1: Reading SEEPROM...done. ahc1: Manual LVD Termination ahc1: BIOS eeprom is present ahc1: Secondary High byte termination Enabled ahc1: Secondary Low byte termination Enabled ahc1: Primary Low Byte termination Enabled ahc1: Primary High Byte termination Enabled ahc1: aic7899 Wide Channel B, SCSI Id=7, 16/255 SCBs ahc1: Downloading Sequencer Program... 399 instructions downloaded fxp0: irq 21 at device 8.0 on pci0 fxp0: Ethernet address 00:c0:4f:a0:46:1a bpf: fxp0 attached isab0: at device 15.0 on pci0 isa0: on isab0 pci0: Unknown PCI ATA controller (vendor=0x1166, dev=0x0211) at 15.1 pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib4: on motherboard pci4: on pcib4 pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 pci_open(1a): mode1res=0x80000000 (0x80000000) pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) pcib12: on motherboard Freeing (NOT implemented) redirected PCI irq 10. found-> vendor=0x1000, dev=0x000b, revid=0x05 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=a, irq=11 map[10]: type 1, range 32, base 00008c00, size 8 map[14]: type 1, range 64, base f9cffc00, size 10 map[1c]: type 1, range 64, base f9cfc000, size 13 Freeing (NOT implemented) redirected PCI irq 5. found-> vendor=0x1000, dev=0x000b, revid=0x05 class=01-00-00, hdrtype=0x00, mfdev=1 subordinatebus=0 secondarybus=0 intpin=b, irq=255 map[10]: type 1, range 32, base 00008800, size 8 map[14]: type 1, range 64, base f9cff800, size 10 map[1c]: type 1, range 64, base f9cfa000, size 13 pci12: on pcib12 sym0: <896> irq 11 at device 12.0 on pci12 sym0: Symbios NVRAM, ID 7, Fast-40, LVD, parity checking sym0: open drain IRQ line driver, using on-chip SRAM sym0: handling phase mismatch from SCRIPTS. sym0: initial SCNTL3/DMODE/DCNTL/CTEST3/4/5 = (hex) 07/4e/a0/01/00/24 sym0: final SCNTL3/DMODE/DCNTL/CTEST3/4/5 = (hex) 07/4e/80/01/08/24 sym0: Delay (GEN=11): 243 msec, 36577 KHz sym0: Delay (GEN=11): 276 msec, 32204 KHz sym0: Delay (GEN=11): 276 msec, 32204 KHz sym0: enabling clock multiplier sym0: Downloading SCSI SCRIPTS. sym1: <896> at device 12.1 on pci12 sym1: failed to allocate IRQ resource Fatal trap 12: page fault while in kernel mode mp_lock = 00000002; cpuid = 0; lapic.id = 03000000 fault virtual address = 0x0 fault code = supervisor read, page not present instruction pointer = 0x8:0xc01f5091 stack pointer = 0x10:0xc034ace8 frame pointer = 0x10:0xc034acf8 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 0 (swapper) interrupt mask = net tty bio cam <- SMP: XXX trap number = 12 panic: page fault mp_lock = 00000002; cpuid = 0; lapic.id = 03000000 Uptime: 0s Automatic reboot in 15 seconds - press a key on the console to abort Rebooting... 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 Wed Feb 9 12: 3: 1 2000 Delivered-To: freebsd-smp@freebsd.org Received: from lor.watermarkgroup.com (lor.watermarkgroup.com [207.202.73.33]) by builder.freebsd.org (Postfix) with ESMTP id 91016424F for ; Wed, 9 Feb 2000 12:02:58 -0800 (PST) Received: (from luoqi@localhost) by lor.watermarkgroup.com (8.8.8/8.8.8) id PAA05819; Wed, 9 Feb 2000 15:01:48 -0500 (EST) (envelope-from luoqi) Date: Wed, 9 Feb 2000 15:01:48 -0500 (EST) From: Luoqi Chen Message-Id: <200002092001.PAA05819@lor.watermarkgroup.com> To: jlemon@flugsvamp.com, smp@FreeBSD.ORG Subject: Re: SMP on Dell Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I'm trying to get a Dell Poweredge up and running SMP and am > having some difficulty. Attached is the mptable from a uni-processor > boot (which works), and the dmesg output from a failed SMP boot. > > As I'm currently lacking enough SMP clues to figure this out, could > someone point me in the right direction? The 'Freeing (not implemented)' > lines look suspicious to me. > -- > Jonathan > Try increasing your NINTR to 32 (the default is 24, but you have two ioapics each with 16 pins). > Programming 16 pins in IOAPIC #0 > Programming 16 pins in IOAPIC #1 > IOAPIC #1 intpin 0 -> irq 2 > IOAPIC #1 intpin 1 -> irq 5 > IOAPIC #1 intpin 2 -> irq 10 > IOAPIC #1 intpin 3 -> irq 11 > IOAPIC #1 intpin 4 -> irq 13 > IOAPIC #1 intpin 5 -> irq 16 > IOAPIC #1 intpin 6 -> irq 17 > IOAPIC #1 intpin 7 -> irq 18 > IOAPIC #1 intpin 8 -> irq 19 > IOAPIC #1 intpin 9 -> irq 20 > IOAPIC #1 intpin 10 -> irq 21 > IOAPIC #1 intpin 11 -> irq 22 > IOAPIC #1 intpin 12 -> irq 23 > I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# > INT conforms conforms 12 12:B 5 13 ^^^^ ^^ > sym1: <896> at device 12.1 on pci12 ^^^^ > sym1: failed to allocate IRQ resource Pin #13 of ioapic #1 is not programmed, that's probably the cause of the crash. -lq To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Feb 9 18:20:13 2000 Delivered-To: freebsd-smp@freebsd.org Received: from prism.flugsvamp.com (cb58709-a.mdsn1.wi.home.com [24.17.241.9]) by builder.freebsd.org (Postfix) with ESMTP id 46579434D for ; Wed, 9 Feb 2000 18:20:11 -0800 (PST) Received: (from jlemon@localhost) by prism.flugsvamp.com (8.9.3/8.9.3) id UAA18320; Wed, 9 Feb 2000 20:21:22 -0600 (CST) (envelope-from jlemon) Date: Wed, 9 Feb 2000 20:21:21 -0600 From: Jonathan Lemon To: Thierry.Herbelot@alcatel.fr Cc: Jonathan Lemon , smp@FreeBSD.ORG Subject: Re: SMP on Dell Message-ID: <20000209202121.A18312@prism.flugsvamp.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0pre2i In-Reply-To: Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, Feb 09, 2000 at 07:44:33PM +0100, Thierry.Herbelot@alcatel.fr wrote: > I'm quite suspicious seeing hints of a non-supported SMP chipset from Renaissance : > pcib4: on motherboard > pci4: on pcib4 > pci_open(1): mode 1 addr port (0x0cf8) is 0x00000000 > pci_open(1a): mode1res=0x80000000 (0x80000000) > pci_cfgcheck: device 0 [class=060000] [hdr=80] is there (id=00081166) > pcib12: on motherboard Well, thanks to the patch that Drew just committed, it's now "supported" by FreeBSD. (in quotes, 'cause SMP isn't working yet) :-) -- Jonathan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Feb 9 20:20:58 2000 Delivered-To: freebsd-smp@freebsd.org Received: from prism.flugsvamp.com (cb58709-a.mdsn1.wi.home.com [24.17.241.9]) by builder.freebsd.org (Postfix) with ESMTP id AA2054381 for ; Wed, 9 Feb 2000 20:20:54 -0800 (PST) Received: (from jlemon@localhost) by prism.flugsvamp.com (8.9.3/8.9.3) id WAA18689; Wed, 9 Feb 2000 22:21:18 -0600 (CST) (envelope-from jlemon) Date: Wed, 9 Feb 2000 22:21:18 -0600 From: Jonathan Lemon To: Luoqi Chen Cc: jlemon@flugsvamp.com, smp@FreeBSD.ORG Subject: Re: SMP on Dell Message-ID: <20000209222118.A18652@prism.flugsvamp.com> References: <200002092001.PAA05819@lor.watermarkgroup.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0pre2i In-Reply-To: <200002092001.PAA05819@lor.watermarkgroup.com> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, Feb 09, 2000 at 03:01:48PM -0500, Luoqi Chen wrote: > > I'm trying to get a Dell Poweredge up and running SMP and am > > having some difficulty. Attached is the mptable from a uni-processor > > boot (which works), and the dmesg output from a failed SMP boot. > > > > As I'm currently lacking enough SMP clues to figure this out, could > > someone point me in the right direction? The 'Freeing (not implemented)' > > lines look suspicious to me. > > -- > > Jonathan > > > Try increasing your NINTR to 32 (the default is 24, but you have two ioapics > each with 16 pins). Actually, I had NINTR set to 46 or so, so this shouldn't have been a problem. What turns out to be the problem is that intpins 13-15 on the APIC are supposed to be set to irq's 24-26. However, irqs > 23 don't appear to exist (there is no ICU vector for them). What would be the correct fix in this case? Try to bump up both APIC_INTMAPSIZE and ICU_LEN (which are both set to 24), or should APIC interrupt number be reprogrammed later to be below 24? Context included below.. > > Programming 16 pins in IOAPIC #0 > > Programming 16 pins in IOAPIC #1 > > IOAPIC #1 intpin 0 -> irq 2 > > IOAPIC #1 intpin 1 -> irq 5 > > IOAPIC #1 intpin 2 -> irq 10 > > IOAPIC #1 intpin 3 -> irq 11 > > IOAPIC #1 intpin 4 -> irq 13 > > IOAPIC #1 intpin 5 -> irq 16 > > IOAPIC #1 intpin 6 -> irq 17 > > IOAPIC #1 intpin 7 -> irq 18 > > IOAPIC #1 intpin 8 -> irq 19 > > IOAPIC #1 intpin 9 -> irq 20 > > IOAPIC #1 intpin 10 -> irq 21 > > IOAPIC #1 intpin 11 -> irq 22 > > IOAPIC #1 intpin 12 -> irq 23 > > > I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# > > INT conforms conforms 12 12:B 5 13 > ^^^^ ^^ > > sym1: <896> at device 12.1 on pci12 > ^^^^ > > sym1: failed to allocate IRQ resource > > Pin #13 of ioapic #1 is not programmed, that's probably the cause of the > crash. -- Jonathan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Feb 10 3: 4:21 2000 Delivered-To: freebsd-smp@freebsd.org Received: from porthos-dmz.barclays.co.uk (mailgate.barclays.co.uk [193.128.3.20]) by builder.freebsd.org (Postfix) with ESMTP id 5D128440F for ; Thu, 10 Feb 2000 03:04:09 -0800 (PST) Received: from athos (actually host athos-dmz.barclays.co.uk) by porthos-dmz.barclays.co.uk with SMTP (XT-PP) with ESMTP; Thu, 10 Feb 2000 11:03:29 +0000 Message-ID: <38A297DF.F27B267D@barclays.co.uk> Date: Thu, 10 Feb 2000 10:50:07 +0000 From: "Paul J. Vickers" X-Mailer: Mozilla 4.7 [en] (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 To: smp@FreeBSD.org Subject: ASUS Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm trying to get FreeBSD 3.4 up on an ASUS P2L97-DS. It boots Ok and runs for a while, then stops dead ?? I enclose the mptable and dmesg out put MPTable, version 2.0.15 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000f6e30 signature: '_MP_' length: 16 bytes version: 1.4 checksum: 0x05 mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x000f6a22 signature: 'PCMP' base table length: 268 version: 1.4 checksum: 0x97 OEM ID: 'OEM00000' Product ID: 'PROD00000000' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 25 local APIC address: 0xfee00000 extended table length: 124 extended table checksum: 184 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 1 0x11 BSP, usable 6 3 4 0x80fbff 0 0x11 AP, usable 6 3 4 0x80fbff -- Bus: Bus ID Type 0 PCI 1 PCI 2 ISA -- I/O APICs: APIC ID Version State Address 2 0x11 usable 0xfec00000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT conforms conforms 2 0 2 0 INT conforms conforms 2 1 2 1 INT conforms conforms 2 0 2 2 INT conforms conforms 2 3 2 3 INT conforms conforms 2 4 2 4 INT conforms conforms 2 6 2 6 INT conforms conforms 2 7 2 7 INT conforms conforms 2 8 2 8 INT conforms conforms 2 9 2 9 INT conforms conforms 2 12 2 12 INT conforms conforms 2 14 2 14 INT conforms conforms 2 15 2 15 INT active-lo level 1 0:A 2 16 INT active-lo level 0 4:D 2 19 INT active-lo level 0 9:A 2 19 INT active-lo level 0 10:A 2 18 INT active-lo level 0 11:A 2 17 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 2 0 255 0 NMI active-hi edge 2 0 255 1 ------------------------------------------------------------------------------- MP Config Extended Table Entries: -- bus ID: 0 address type: I/O address address base: 0x0 address range: 0x10000 -- bus ID: 0 address type: memory address address base: 0x8000000 address range: 0xdae80000 -- bus ID: 0 address type: prefetch address address base: 0xe2e80000 address range: 0x5180000 -- bus ID: 0 address type: memory address address base: 0xe8000000 address range: 0x18000000 -- bus ID: 0 address type: memory address address base: 0xa0000 address range: 0x20000 -- bus ID: 2 bus info: 0x01 parent bus ID: 0-- bus ID: 0 address modifier: add predefined range: 0x00000000-- bus ID: 0 address modifier: add predefined range: 0x00000001 ------------------------------------------------------------------------------- # SMP kernel config file options: # Required: options SMP # Symmetric MultiProcessor Kernel options APIC_IO # Symmetric (APIC) I/O # Optional (built-in defaults will work in most cases): options NCPU=2 # number of CPUs options NBUS=3 # number of busses options NAPIC=1 # number of IO APICs options NINTR=24 # number of INTs =============================================================================== Copyright (c) 1992-1999 FreeBSD Inc. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 3.4-RELEASE #2: Wed Feb 9 16:45:08 GMT 2000 root@erin.ipv6.stedman.co.uk:/usr/kame/freebsd3/sys/compile/IPV6 Timecounter "i8254" frequency 1193182 Hz CPU: Pentium II (686-class CPU) Origin = "GenuineIntel" Id = 0x634 Stepping = 4 Features=0x80fbff real memory = 134217728 (131072K bytes) avail memory = 126832640 (123860K bytes) Programming 24 pins in IOAPIC #0 FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 1, version: 0x00040011, at 0xfee00000 cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 io0 (APIC): apic id: 2, version: 0x00170011, at 0xfec00000 Preloaded elf kernel "kernel" at 0xc037b000. Pentium Pro MTRR support enabled altq: major number is 96 Probing for devices on PCI bus 0: chip0: rev 0x03 on pci0.0.0 chip1: rev 0x03 on pci0.1.0 chip2: rev 0x01 on pci0.4.0 ide_pci0: rev 0x01 on pci0.4.1 chip3: rev 0x01 on pci0.4.3 ahc0: rev 0x00 int a irq 19 on pci0.9.0 ahc0: aic7880 Wide Channel A, SCSI Id=7, 16/255 SCBs pn0: <82c169 PNIC 10/100BaseTX> rev 0x21 int a irq 18 on pci0.10.0 pn0: Ethernet address: 00:a0:cc:39:e7:8a pn0: autoneg complete, link status good (full-duplex, 100Mbps) pn0: supplying EUI64: 00:a0:cc:ff:fe:39:e7:8a de0: rev 0x12 int a irq 17 on pci0.11.0 de0: DEC DE500-XA 21140 [10-100Mb/s] pass 1.2 de0: address 00:00:f8:30:65:41 Probing for devices on PCI bus 1: vga0: rev 0x02 int a irq 16 on pci1.0.0 Probing for PnP devices: Probing for devices on the ISA bus: sc0 on isa sc0: VGA color <16 virtual consoles, flags=0x0> atkbdc0 at 0x60-0x6f on motherboard atkbd0 irq 1 on isa psm0 irq 12 on isa psm0: model Generic PS/2 mouse, device ID 0 sio0 at 0x3f8-0x3ff irq 4 flags 0x10 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: FIFO enabled, 8 bytes threshold fd0: 1.44MB 3.5in wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (wd0): wd0: 9671MB (19807200 sectors), 19650 cyls, 16 heads, 63 S/T, 512 B/S wdc1 at 0x170-0x177 irq 15 on isa wdc1: unit 0 (atapi): , removable, accel, dma, iordy acd0: drive speed 1031 - 5843KB/sec, 128KB cache acd0: supported read types: CD-R, CD-RW, CD-DA, packet track acd0: Audio: play, 255 volume levels acd0: Mechanism: ejectable tray acd0: Medium: CD-ROM 120mm data disc loaded, unlocked wdc1: unit 1 (atapi): , removable, intr, iordis wfd0: medium type unknown (no disk) wfd0: buggy Zip drive, 64-block transfer limit set ppc0 at 0x378 irq 7 flags 0x40 on isa ppc0: SMC-like chipset (ECP/EPP/PS2/NIBBLE) in COMPATIBLE mode ppc0: FIFO with 16/16/9 bytes threshold lpt0: on ppbus 0 lpt0: Interrupt-driven port ppi0: on ppbus 0 plip0: on ppbus 0 vga0 at 0x3b0-0x3df maddr 0xa0000 msize 131072 on isa npx0 on motherboard npx0: INT 16 interface APIC_IO: Testing 8254 interrupt delivery APIC_IO: routing 8254 via pin 2 IPsec: Initialized Security Association Processing. Waiting 2 seconds for SCSI devices to settle SMP: AP CPU #1 Launched! de0: enabling 10baseT port sa0 at ahc0 bus 0 target 2 lun 0 sa0: Removable Sequential Access SCSI-2 device sa0: 10.000MB/s transfers (10.000MHz, offset 15) sa1 at ahc0 bus 0 target 5 lun 0 sa1: Removable Sequential Access SCSI-2 device sa1: 4.032MB/s transfers (4.032MHz, offset 11) Internet communications are not secure and therefore the Barclays Group does not accept legal responsibility for the contents of this message. Any views or opinions presented are solely those of the author and do not necessarily represent those of the Barclays Group. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 12:37:47 2000 Delivered-To: freebsd-smp@freebsd.org Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [206.168.13.65]) by builder.freebsd.org (Postfix) with ESMTP id 18DA6422E for ; Sun, 13 Feb 2000 12:37:40 -0800 (PST) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.9.3/8.9.3) with ESMTP id NAA44926 for ; Sun, 13 Feb 2000 13:37:46 -0700 (MST) (envelope-from fbsd@Ilsa.StevesCafe.com) Message-Id: <200002132037.NAA44926@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0.2 2/24/98 From: Steve Passe To: freebsd-smp@FreeBSD.ORG Subject: current hardware choices. Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Date: Sun, 13 Feb 2000 13:37:46 -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, I need to spend some cash b4 the end of the first tax quarter, so... What makes the most sense for a quad SMP box today? = --- The Intel SC450NX is available for around $3500, Pentium=AE II Xeon 400MHz MMX w/1024K 1MB cache - OEM PULLS are availabl= e for $200, New boxed PIII Xeon 500MHz w/512K cache go for around $950, What would give overall better performance the extra cache, or the extra MHz? --- New boxed PIII Xeon 733MHz w/512K cache coppermine go for $750, will any coppermine capable quad boards be available anytime in the forseeable future? Do they even make sense for a quad system, eg. the very small cache? -- Steve Passe | powered by = smp@csn.net | Symmetric MultiProcessor FreeBSD To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 13:23:49 2000 Delivered-To: freebsd-smp@freebsd.org Received: from panzer.kdm.org (panzer.kdm.org [216.160.178.169]) by builder.freebsd.org (Postfix) with ESMTP id 67E87406C for ; Sun, 13 Feb 2000 13:23:45 -0800 (PST) Received: (from ken@localhost) by panzer.kdm.org (8.9.3/8.9.1) id OAA08070; Sun, 13 Feb 2000 14:23:40 -0700 (MST) (envelope-from ken) Date: Sun, 13 Feb 2000 14:23:40 -0700 From: "Kenneth D. Merry" To: Steve Passe Cc: freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. Message-ID: <20000213142340.A7998@panzer.kdm.org> References: <200002132037.NAA44926@Ilsa.StevesCafe.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Mailer: Mutt 1.0i In-Reply-To: <200002132037.NAA44926@Ilsa.StevesCafe.com>; from smp@csn.net on Sun, Feb 13, 2000 at 01:37:46PM -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Feb 13, 2000 at 13:37:46 -0700, Steve Passe wrote: > Hi, > > I need to spend some cash b4 the end of the first tax quarter, so... > > What makes the most sense for a quad SMP box today? > > --- > The Intel SC450NX is available for around $3500, > PentiumŪ II Xeon 400MHz MMX w/1024K 1MB cache - OEM PULLS are available > for $200, > New boxed PIII Xeon 500MHz w/512K cache go for around $950, > > What would give overall better performance the extra cache, or the > extra MHz? > > --- > New boxed PIII Xeon 733MHz w/512K cache coppermine go for $750, > will any coppermine capable quad boards be available anytime in the > forseeable future? > > Do they even make sense for a quad system, eg. the very small cache? The Coppermine Xeons have 256K cache, not 512K, AFAIK. I don't think chips with 256K cache would perform all that well in a 4-way SMP setup. In any case, the new Coppermine Xeons only support 2-way SMP, so you'll have to go with a 550MHz or slower Xeon to do 4-way SMP. If you really want to spend some money, you should go for an 8-way box based on the Profusion chipset. :) IIRC, the Intel Profusion motherboard has something like 10 64bit PCI slots, 4 of them at 66MHz. :) (A month or two ago, I had fun with the configuration tool on Dell's web site, and managed to load enough options on one of their 8-way servers to drive the price up to ~$110,000.) Ken -- Kenneth Merry ken@kdm.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 13:28:53 2000 Delivered-To: freebsd-smp@freebsd.org Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [206.168.13.65]) by builder.freebsd.org (Postfix) with ESMTP id 8091A3E35 for ; Sun, 13 Feb 2000 13:28:46 -0800 (PST) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.9.3/8.9.3) with ESMTP id OAA45524; Sun, 13 Feb 2000 14:28:42 -0700 (MST) (envelope-from smp@Ilsa.StevesCafe.com) Message-Id: <200002132128.OAA45524@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0.2 2/24/98 From: Steve Passe To: "Kenneth D. Merry" Cc: freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. In-reply-to: Your message of "Sun, 13 Feb 2000 14:23:40 MST." <20000213142340.A7998@panzer.kdm.org> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Date: Sun, 13 Feb 2000 14:28:42 -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > On Sun, Feb 13, 2000 at 13:37:46 -0700, Steve Passe wrote: > ... > > --- > > New boxed PIII Xeon 733MHz w/512K cache coppermine go for $750, > > will any coppermine capable quad boards be available anytime in the= > > forseeable future? > > = > > Do they even make sense for a quad system, eg. the very small cache?= > = > The Coppermine Xeons have 256K cache, not 512K, AFAIK. I don't think c= hips cut-n-paste error, I knew that! > with 256K cache would perform all that well in a 4-way SMP setup. In a= ny > case, the new Coppermine Xeons only support 2-way SMP, so you'll have t= o go > with a 550MHz or slower Xeon to do 4-way SMP. > = > If you really want to spend some money, you should go for an 8-way box > based on the Profusion chipset. :) IIRC, the Intel Profusion motherboa= rd > has something like 10 64bit PCI slots, 4 of them at 66MHz. :) (A month= or > two ago, I had fun with the configuration tool on Dell's web site, and > managed to load enough options on one of their 8-way servers to drive t= he > price up to ~$110,000.) I was just looking at that server on the intel site: OCPRF100, but I couldn't find it listed on any of the web price engines. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 13:31:38 2000 Delivered-To: freebsd-smp@freebsd.org Received: from panzer.kdm.org (panzer.kdm.org [216.160.178.169]) by builder.freebsd.org (Postfix) with ESMTP id 3A2ED4292 for ; Sun, 13 Feb 2000 13:31:34 -0800 (PST) Received: (from ken@localhost) by panzer.kdm.org (8.9.3/8.9.1) id OAA08192; Sun, 13 Feb 2000 14:31:35 -0700 (MST) (envelope-from ken) Date: Sun, 13 Feb 2000 14:31:35 -0700 From: "Kenneth D. Merry" To: Steve Passe Cc: freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. Message-ID: <20000213143135.A8161@panzer.kdm.org> References: <20000213142340.A7998@panzer.kdm.org> <200002132128.OAA45524@Ilsa.StevesCafe.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <200002132128.OAA45524@Ilsa.StevesCafe.com>; from smp@csn.net on Sun, Feb 13, 2000 at 02:28:42PM -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Feb 13, 2000 at 14:28:42 -0700, Steve Passe wrote: > > On Sun, Feb 13, 2000 at 13:37:46 -0700, Steve Passe wrote: > > ... > > > --- > > > New boxed PIII Xeon 733MHz w/512K cache coppermine go for $750, > > > will any coppermine capable quad boards be available anytime in the > > > forseeable future? > > > > > > Do they even make sense for a quad system, eg. the very small cache? > > > > The Coppermine Xeons have 256K cache, not 512K, AFAIK. I don't think chips > cut-n-paste error, I knew that! Heh. :) > > with 256K cache would perform all that well in a 4-way SMP setup. In any > > case, the new Coppermine Xeons only support 2-way SMP, so you'll have to go > > with a 550MHz or slower Xeon to do 4-way SMP. > > > > If you really want to spend some money, you should go for an 8-way box > > based on the Profusion chipset. :) IIRC, the Intel Profusion motherboard > > has something like 10 64bit PCI slots, 4 of them at 66MHz. :) (A month or > > two ago, I had fun with the configuration tool on Dell's web site, and > > managed to load enough options on one of their 8-way servers to drive the > > price up to ~$110,000.) > > I was just looking at that server on the intel site: OCPRF100, but > I couldn't find it listed on any of the web price engines. Yeah, I looked around for a Profusion server from some of the clone-type places, but I wasn't able to find it. All I found were the 4-way SMP servers. It looks like you have to go with one of the big vendors to get a Profusion box. Ken -- Kenneth Merry ken@kdm.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 13:52:40 2000 Delivered-To: freebsd-smp@freebsd.org Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [206.168.13.65]) by builder.freebsd.org (Postfix) with ESMTP id DEE674683 for ; Sun, 13 Feb 2000 13:52:31 -0800 (PST) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.9.3/8.9.3) with ESMTP id OAA45725; Sun, 13 Feb 2000 14:52:37 -0700 (MST) (envelope-from smp@Ilsa.StevesCafe.com) Message-Id: <200002132152.OAA45725@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0.2 2/24/98 From: Steve Passe To: "Kenneth D. Merry" Cc: freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. In-reply-to: Your message of "Sun, 13 Feb 2000 14:31:35 MST." <20000213143135.A8161@panzer.kdm.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 13 Feb 2000 14:52:37 -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > On Sun, Feb 13, 2000 at 14:28:42 -0700, Steve Passe wrote: > ... > > I was just looking at that server on the intel site: OCPRF100, but > > I couldn't find it listed on any of the web price engines. > > Yeah, I looked around for a Profusion server from some of the clone-type > places, but I wasn't able to find it. All I found were the 4-way SMP > servers. It looks like you have to go with one of the big vendors to get > a Profusion box. I just priced out the Dell version: $51K for 8 550/1024K! So I guess its an SC450NX for me... Now to decide just home much ransom to pay intel for increased cache size. Has anyone seen any benchmarks that quantify the effects of different L2 cache sizes in SMP systems? -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 14:26:54 2000 Delivered-To: freebsd-smp@freebsd.org Received: from synthcom.com (ip67.usw3.rb1.pdx.nwlink.com [207.202.141.67]) by builder.freebsd.org (Postfix) with ESMTP id 05B3B4373 for ; Sun, 13 Feb 2000 14:26:52 -0800 (PST) Received: from beacon.synthcom.com (beacon.synthcom.com [207.202.141.81]) by synthcom.com (8.9.3/8.9.3) with ESMTP id OAA49418 for ; Sun, 13 Feb 2000 14:40:50 -0800 (PST) (envelope-from neil@beacon.synthcom.com) Date: Sun, 13 Feb 2000 14:40:50 -0800 (PST) From: Neil Bradley To: freebsd-smp@freebsd.org Subject: 4.0 SMP Problem? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm not sure what exactly is going on, but I've gotten three kernel faults while untarring a file when FreeBSD 4.0 current is running in SMP mode. It doesn't always happen, and appears to be fairly unpredictable. If I disable SMP, I don't get the kernel faults. Alternately, my FreeBSD 3.4 operation in SMP mode is dead solid. I'd like to help gather info on this problem for the appropriate people. To whom should I speak to about this? What other details can I give to assist? Thanks! -->Neil ------------------------------------------------------------------------------- Neil Bradley Synthcom home : http://www.synthcom.com Synthcom Systems, Inc. ICQ # 29402898 "Hookt on Fonix" Workt four me! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 21:44: 4 2000 Delivered-To: freebsd-smp@freebsd.org Received: from implode.root.com (root.com [209.102.106.178]) by builder.freebsd.org (Postfix) with ESMTP id CC47548EB for ; Sun, 13 Feb 2000 21:44:01 -0800 (PST) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.8/8.8.5) with ESMTP id VAA11727; Sun, 13 Feb 2000 21:41:45 -0800 (PST) Message-Id: <200002140541.VAA11727@implode.root.com> To: Steve Passe Cc: "Kenneth D. Merry" , freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. In-reply-to: Your message of "Sun, 13 Feb 2000 14:52:37 MST." <200002132152.OAA45725@Ilsa.StevesCafe.com> From: David Greenman Reply-To: dg@root.com Date: Sun, 13 Feb 2000 21:41:44 -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >> On Sun, Feb 13, 2000 at 14:28:42 -0700, Steve Passe wrote: >> ... >> > I was just looking at that server on the intel site: OCPRF100, but >> > I couldn't find it listed on any of the web price engines. >> >> Yeah, I looked around for a Profusion server from some of the clone-type >> places, but I wasn't able to find it. All I found were the 4-way SMP >> servers. It looks like you have to go with one of the big vendors to get >> a Profusion box. > >I just priced out the Dell version: $51K for 8 550/1024K! >So I guess its an SC450NX for me... > >Now to decide just home much ransom to pay intel for increased cache size. >Has anyone seen any benchmarks that quantify the effects of different >L2 cache sizes in SMP systems? The OCPRF100 is sold by TeraSolutions as a model TS-5000. I don't know what kind of a deal we can for you or if the pricing will be better than the big vendors, but we'd be happy to provide a price quote for whatever configuration you specify. -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project - http://www.freebsd.org Creator of high-performance Internet servers - http://www.terasolutions.com Pave the road of life with opportunities. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 21:54:42 2000 Delivered-To: freebsd-smp@freebsd.org Received: from panzer.kdm.org (panzer.kdm.org [216.160.178.169]) by builder.freebsd.org (Postfix) with ESMTP id 5928248CC for ; Sun, 13 Feb 2000 21:54:39 -0800 (PST) Received: (from ken@localhost) by panzer.kdm.org (8.9.3/8.9.1) id WAA11073; Sun, 13 Feb 2000 22:54:42 -0700 (MST) (envelope-from ken) Date: Sun, 13 Feb 2000 22:54:42 -0700 From: "Kenneth D. Merry" To: David Greenman Cc: Steve Passe , freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. Message-ID: <20000213225442.A11028@panzer.kdm.org> References: <200002132152.OAA45725@Ilsa.StevesCafe.com> <200002140541.VAA11727@implode.root.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <200002140541.VAA11727@implode.root.com>; from dg@root.com on Sun, Feb 13, 2000 at 09:41:44PM -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Feb 13, 2000 at 21:41:44 -0800, David Greenman wrote: > >> On Sun, Feb 13, 2000 at 14:28:42 -0700, Steve Passe wrote: > >> ... > >> > I was just looking at that server on the intel site: OCPRF100, but > >> > I couldn't find it listed on any of the web price engines. > >> > >> Yeah, I looked around for a Profusion server from some of the clone-type > >> places, but I wasn't able to find it. All I found were the 4-way SMP > >> servers. It looks like you have to go with one of the big vendors to get > >> a Profusion box. > > > >I just priced out the Dell version: $51K for 8 550/1024K! > >So I guess its an SC450NX for me... > > > >Now to decide just home much ransom to pay intel for increased cache size. > >Has anyone seen any benchmarks that quantify the effects of different > >L2 cache sizes in SMP systems? > > The OCPRF100 is sold by TeraSolutions as a model TS-5000. I don't know what > kind of a deal we can for you or if the pricing will be better than the big > vendors, but we'd be happy to provide a price quote for whatever configuration > you specify. Will FreeBSD actually run on a Profusion board? Has anyone tried it? Ken -- Kenneth Merry ken@kdm.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 22:38:58 2000 Delivered-To: freebsd-smp@freebsd.org Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [206.168.13.65]) by builder.freebsd.org (Postfix) with ESMTP id 146583E5F for ; Sun, 13 Feb 2000 22:38:56 -0800 (PST) Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.9.3/8.9.3) with ESMTP id XAA07584; Sun, 13 Feb 2000 23:39:00 -0700 (MST) (envelope-from smp@Ilsa.StevesCafe.com) Message-Id: <200002140639.XAA07584@Ilsa.StevesCafe.com> X-Mailer: exmh version 2.0.2 2/24/98 From: Steve Passe To: "Kenneth D. Merry" Cc: David Greenman , freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. In-reply-to: Your message of "Sun, 13 Feb 2000 22:54:42 MST." <20000213225442.A11028@panzer.kdm.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 13 Feb 2000 23:39:00 -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, --- > On Sun, Feb 13, 2000 at 21:41:44 -0800, David Greenman wrote: > ... > > The OCPRF100 is sold by TeraSolutions as a model TS-5000. I don't know what do they have a web site? --- > Will FreeBSD actually run on a Profusion board? Has anyone tried it? I asked myself that same question when reading about the "cross-switch" bus arch. it impiments. The intel web page does claim that it is MP spec 1.1 and 1.4 complient... --- ps: evidently my mail is getting out to freeBSD-smp, but I haven't seen any of it get to my freeBSD-smp maillist. I actually haven't seen any mail from that list for months, just assummed it was a dead list now that SMP is mainstream... -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 22:43:53 2000 Delivered-To: freebsd-smp@freebsd.org Received: from panzer.kdm.org (panzer.kdm.org [216.160.178.169]) by builder.freebsd.org (Postfix) with ESMTP id B06173DE4 for ; Sun, 13 Feb 2000 22:43:50 -0800 (PST) Received: (from ken@localhost) by panzer.kdm.org (8.9.3/8.9.1) id XAA36903; Sun, 13 Feb 2000 23:43:58 -0700 (MST) (envelope-from ken) Date: Sun, 13 Feb 2000 23:43:58 -0700 From: "Kenneth D. Merry" To: Steve Passe Cc: David Greenman , freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. Message-ID: <20000213234358.A36878@panzer.kdm.org> References: <20000213225442.A11028@panzer.kdm.org> <200002140639.XAA07584@Ilsa.StevesCafe.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <200002140639.XAA07584@Ilsa.StevesCafe.com>; from smp@csn.net on Sun, Feb 13, 2000 at 11:39:00PM -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Feb 13, 2000 at 23:39:00 -0700, Steve Passe wrote: > Hi, > > --- > > On Sun, Feb 13, 2000 at 21:41:44 -0800, David Greenman wrote: > > ... > > > The OCPRF100 is sold by TeraSolutions as a model TS-5000. I don't know what > > do they have a web site? http://www.terasolutions.com/ > --- > > Will FreeBSD actually run on a Profusion board? Has anyone tried it? > > I asked myself that same question when reading about the "cross-switch" > bus arch. it impiments. The intel web page does claim that it is MP spec 1.1 > and 1.4 complient... > > --- > ps: > evidently my mail is getting out to freeBSD-smp, but I haven't seen > any of it get to my freeBSD-smp maillist. I actually haven't seen any > mail from that list for months, just assummed it was a dead list now that > SMP is mainstream... You may have been unsubscribed if your mail was bouncing or something. You might want to see if you're still subscribed. (I think sending the 'which' command to majordomo@FreeBSD.ORG might do the trick.) Ken -- Kenneth Merry ken@kdm.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 22:45:48 2000 Delivered-To: freebsd-smp@freebsd.org Received: from milhouse.cnbinc.com (martin2.lnk.telstra.net [139.130.137.91]) by builder.freebsd.org (Postfix) with ESMTP id 66D0E3F6F for ; Sun, 13 Feb 2000 22:45:38 -0800 (PST) Received: from bender (bender.cnbinc.com [203.38.182.20]) by milhouse.cnbinc.com (8.9.3/8.9.3) with SMTP id RAA30028; Mon, 14 Feb 2000 17:12:54 +1030 (CST) (envelope-from diskiller@cnbinc.com) Message-ID: <000b01bf76b6$b7d24740$14b626cb@bender.cnbinc.com> From: "Martin Minkus" To: "Neil Bradley" Cc: References: Subject: Re: 4.0 SMP Problem? Date: Mon, 14 Feb 2000 17:12:51 +1030 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 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I'm not sure what exactly is going on, but I've gotten three kernel faults > while untarring a file when FreeBSD 4.0 current is running in SMP mode. It > doesn't always happen, and appears to be fairly unpredictable. > > If I disable SMP, I don't get the kernel faults. Alternately, my FreeBSD > 3.4 operation in SMP mode is dead solid. > > I'd like to help gather info on this problem for the appropriate people. > To whom should I speak to about this? What other details can I give to > assist? Your not the only one. I had 4.0 panic, and crash time and time again. Sometimes it'd panic before it could boot all the way up. It also liked to spontaneously reboot quite often. Disabling softupdates got rid of alot of the crashes, and the rest were due to SMP. SMP alone, and softupdate alone, and everything else, is broken on 4.0. I cvsup'd a few days ago, so it was a recent current. No, i am not going to try it again, this is a server, and i can't afford anymore downtime. Once i get another machine i can play on, i would love to move my 4.0-current source tree over, cvsup to update it, and make world to try it out and play some more. But its not possible right now. But then again, 3.4 with SMP + Softupdates is cause for trouble too and causes constant panics for me. Disable softupdates, and all is sweet again. I use softupdates on a non-smp system (also 3.4) and all is sweet aswell. So SMP+Softupdates on 3.4 is seriously broken. 3.4 + SMP + No Softupdates + Async is damn fast, and reliable. So thats what i use. martin. -- diskiller@cnbinc.com Linux/FreeBSD hacker Please use PGP: http://www.diskiller.net/pgp.html To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 23:10:30 2000 Delivered-To: freebsd-smp@freebsd.org Received: from implode.root.com (root.com [209.102.106.178]) by builder.freebsd.org (Postfix) with ESMTP id BA0264024 for ; Sun, 13 Feb 2000 23:10:24 -0800 (PST) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.8/8.8.5) with ESMTP id XAA11927; Sun, 13 Feb 2000 23:08:26 -0800 (PST) Message-Id: <200002140708.XAA11927@implode.root.com> To: "Kenneth D. Merry" Cc: Steve Passe , freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. In-reply-to: Your message of "Sun, 13 Feb 2000 22:54:42 MST." <20000213225442.A11028@panzer.kdm.org> From: David Greenman Reply-To: dg@root.com Date: Sun, 13 Feb 2000 23:08:26 -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >> The OCPRF100 is sold by TeraSolutions as a model TS-5000. I don't know what >> kind of a deal we can for you or if the pricing will be better than the big >> vendors, but we'd be happy to provide a price quote for whatever configuration >> you specify. > >Will FreeBSD actually run on a Profusion board? Has anyone tried it? The OCPRF100 just started shipping from Intel recently and I haven't seen or tested one myself. All we've done at TeraSolutions so far is assign a TSI model number to it. Of course we've shipped many SC450NX and AC450NX systems and they mostly work with FreeBSD except for the onboard SCSI (we ship those systems with PCI Adaptec SCSI controllers to work around this). On the other hand, I'm sure I would have heard something if FreeBSD didn't work with the OCPRF100 since I talk to people at Intel in the high-end server division and they do test FreeBSD on their servers. -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project - http://www.freebsd.org Creator of high-performance Internet servers - http://www.terasolutions.com Pave the road of life with opportunities. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Feb 13 23:14:16 2000 Delivered-To: freebsd-smp@freebsd.org Received: from implode.root.com (root.com [209.102.106.178]) by builder.freebsd.org (Postfix) with ESMTP id 7A58F3DBA for ; Sun, 13 Feb 2000 23:14:11 -0800 (PST) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.8/8.8.5) with ESMTP id XAA11954; Sun, 13 Feb 2000 23:12:56 -0800 (PST) Message-Id: <200002140712.XAA11954@implode.root.com> To: Steve Passe Cc: "Kenneth D. Merry" , freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. In-reply-to: Your message of "Sun, 13 Feb 2000 23:39:00 MST." <200002140639.XAA07584@Ilsa.StevesCafe.com> From: David Greenman Reply-To: dg@root.com Date: Sun, 13 Feb 2000 23:12:56 -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >> On Sun, Feb 13, 2000 at 21:41:44 -0800, David Greenman wrote: >> ... >> > The OCPRF100 is sold by TeraSolutions as a model TS-5000. I don't know what > >do they have a web site? TeraSolutions is my company, and we do have a web site at http://www.terasolutions.com. The site is pretty lame, though, and doesn't have a whole lot of product information online. Intel, however, does have some information online about their servers at developer.intel.com. -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project - http://www.freebsd.org Creator of high-performance Internet servers - http://www.terasolutions.com Pave the road of life with opportunities. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 14 8: 1:52 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mail.enteract.com (mail.enteract.com [207.229.143.33]) by builder.freebsd.org (Postfix) with ESMTP id B3DF8498C for ; Mon, 14 Feb 2000 08:01:41 -0800 (PST) Received: from shell-2.enteract.com (jrs@shell-2.enteract.com [207.229.143.41]) by mail.enteract.com (8.9.3/8.9.3) with SMTP id KAA80384; Mon, 14 Feb 2000 10:01:30 -0600 (CST) (envelope-from jrs@enteract.com) Date: Mon, 14 Feb 2000 10:01:30 -0600 (CST) From: John Sconiers To: Martin Minkus Cc: Neil Bradley , freebsd-smp@freebsd.org Subject: Re: 4.0 SMP Problem? In-Reply-To: <000b01bf76b6$b7d24740$14b626cb@bender.cnbinc.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm using a pretty current SMP system and I'm not having these problems. In my kernel I have but not using soffupdates. The only "big" problem I saw was with sendmail core dumping. > Your not the only one. I had 4.0 panic, and crash time and time again. > Sometimes it'd panic before it could boot all the way up. It also liked > to spontaneously reboot quite often. > Disabling softupdates got rid of alot of the crashes, and the rest were > due to SMP. > SMP alone, and softupdate alone, and everything else, is broken on 4.0. > I cvsup'd a few days ago, so it was a recent current. No, i am not > going to try it again, this is a server, and i can't afford anymore > downtime. > Once i get another machine i can play on, i would love to move my > 4.0-current source tree over, cvsup to update it, and make world to > try it out and play some more. But its not possible right now. > But then again, 3.4 with SMP + Softupdates is cause for trouble too > and causes constant panics for me. Disable softupdates, and all is sweet > again. I use softupdates on a non-smp system (also 3.4) and all is > sweet aswell. So SMP+Softupdates on 3.4 is seriously broken. > 3.4 + SMP + No Softupdates + Async is damn fast, and reliable. So > thats what i use. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 14 8:18:48 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mail.wolves.k12.mo.us (mail.wolves.k12.mo.us [207.160.214.1]) by builder.freebsd.org (Postfix) with ESMTP id 9945F3F8C for ; Mon, 14 Feb 2000 08:18:39 -0800 (PST) Received: from mail.wolves.k12.mo.us (cdillon@mail.wolves.k12.mo.us [207.160.214.1]) by mail.wolves.k12.mo.us (8.9.3/8.9.3) with ESMTP id KAA37537; Mon, 14 Feb 2000 10:18:29 -0600 (CST) (envelope-from cdillon@wolves.k12.mo.us) Date: Mon, 14 Feb 2000 10:18:29 -0600 (CST) From: Chris Dillon To: David Greenman Cc: "Kenneth D. Merry" , Steve Passe , freebsd-smp@FreeBSD.ORG Subject: Re: current hardware choices. In-Reply-To: <200002140708.XAA11927@implode.root.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, 13 Feb 2000, David Greenman wrote: > >> The OCPRF100 is sold by TeraSolutions as a model TS-5000. I don't know what > >> kind of a deal we can for you or if the pricing will be better than the big > >> vendors, but we'd be happy to provide a price quote for whatever configuration > >> you specify. > > > >Will FreeBSD actually run on a Profusion board? Has anyone tried it? > > The OCPRF100 just started shipping from Intel recently and I haven't seen > or tested one myself. All we've done at TeraSolutions so far is assign a TSI > model number to it. Of course we've shipped many SC450NX and AC450NX systems > and they mostly work with FreeBSD except for the onboard SCSI (we ship those > systems with PCI Adaptec SCSI controllers to work around this). Hmm... According to Intel's specs, the SC450NX uses the Symbios Logic 53C896 for two wide-ultra2 SCSI channels, and a 53C810AE for a single narrow channel. Have you tried the newer sym driver to see if it works with those? I'm curious, since I'm thinking of using that board for some servers as well. I'm not sure why the ncr driver wouldn't work with those either, though. > On the other hand, I'm sure I would have heard something if FreeBSD didn't > work with the OCPRF100 since I talk to people at Intel in the high-end server > division and they do test FreeBSD on their servers. Thats always good to know. :-) -- Chris Dillon - cdillon@wolves.k12.mo.us - cdillon@inter-linc.net FreeBSD: The fastest and most stable server OS on the planet. For Intel x86 and Alpha architectures. ( 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 Feb 14 9: 2:28 2000 Delivered-To: freebsd-smp@freebsd.org Received: from post.mail.nl.demon.net (post-11.mail.nl.demon.net [194.159.73.21]) by builder.freebsd.org (Postfix) with ESMTP id D629D46C0 for ; Mon, 14 Feb 2000 09:02:22 -0800 (PST) Received: from [212.238.105.241] (helo=propro) by post.mail.nl.demon.net with esmtp (Exim 2.12 #1) id 12KOtF-000D61-00; Mon, 14 Feb 2000 17:02:25 +0000 Date: Mon, 14 Feb 2000 18:02:24 +0100 (CET) From: Marc Schneiders To: Martin Minkus Cc: Neil Bradley , freebsd-smp@freebsd.org Subject: Re: 4.0 SMP Problem? In-Reply-To: <000b01bf76b6$b7d24740$14b626cb@bender.cnbinc.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 14 Feb 2000, Martin Minkus wrote: [...]> > SMP alone, and softupdate alone, and everything else, is broken on 4.0. > I cvsup'd a few days ago, so it was a recent current. I don't know about softupdates, which I do not use. SMP alone is *not* broken on 4.0. It never was, not even during the recent (x)install and lib problems. -- Marc Schneiders marc@venster.nl http://zelf.net marc@oldserver.demon.nl propro 5:18pm up 6 days, 19:29, load average: 2.00 2.03 1.84 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 14 9:42:15 2000 Delivered-To: freebsd-smp@freebsd.org Received: from verdi.nethelp.no (verdi.nethelp.no [158.36.41.162]) by builder.freebsd.org (Postfix) with SMTP id 164384261 for ; Mon, 14 Feb 2000 09:42:07 -0800 (PST) Received: (qmail 28106 invoked by uid 1001); 14 Feb 2000 17:42:18 +0000 (GMT) To: marc@oldserver.demon.nl Cc: diskiller@cnbinc.com, neil@synthcom.com, freebsd-smp@freebsd.org Subject: Re: 4.0 SMP Problem? From: sthaug@nethelp.no In-Reply-To: Your message of "Mon, 14 Feb 2000 18:02:24 +0100 (CET)" References: X-Mailer: Mew version 1.05+ on Emacs 19.34.2 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Mon, 14 Feb 2000 18:42:18 +0100 Message-ID: <28104.950550138@verdi.nethelp.no> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > SMP alone, and softupdate alone, and everything else, is broken on 4.0. > > I cvsup'd a few days ago, so it was a recent current. > > I don't know about softupdates, which I do not use. SMP alone is *not* > broken on 4.0. It never was, not even during the recent (x)install > and lib problems. I'm using SMP *and* softupdates on a system here. It survives multiple buildworlds just fine - so I'd say things are working pretty well. Steinar Haug, Nethelp consulting, sthaug@nethelp.no To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 14 12:44:30 2000 Delivered-To: freebsd-smp@freebsd.org Received: from zippy.cdrom.com (zippy.cdrom.com [204.216.27.228]) by builder.freebsd.org (Postfix) with ESMTP id 96C663F4D for ; Mon, 14 Feb 2000 12:44:28 -0800 (PST) Received: from zippy.cdrom.com (jkh@localhost [127.0.0.1]) by zippy.cdrom.com (8.9.3/8.9.3) with ESMTP id MAA48374; Mon, 14 Feb 2000 12:43:11 -0800 (PST) (envelope-from jkh@zippy.cdrom.com) To: sthaug@nethelp.no Cc: marc@oldserver.demon.nl, diskiller@cnbinc.com, neil@synthcom.com, freebsd-smp@FreeBSD.ORG Subject: Re: 4.0 SMP Problem? In-reply-to: Your message of "Mon, 14 Feb 2000 18:42:18 +0100." <28104.950550138@verdi.nethelp.no> Date: Mon, 14 Feb 2000 12:43:09 -0800 Message-ID: <48371.950560989@zippy.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I have to echo these sentiments; I've been running softupdates on my SMP box and building frequent releases on it, among other disk-intensive exercises, and it works flawlessly. Has for months, in fact, despite known-bugs getting fixed in soft updates during that time. - Jordan > > > SMP alone, and softupdate alone, and everything else, is broken on 4.0. > > > I cvsup'd a few days ago, so it was a recent current. > > > > I don't know about softupdates, which I do not use. SMP alone is *not* > > broken on 4.0. It never was, not even during the recent (x)install > > and lib problems. > > I'm using SMP *and* softupdates on a system here. It survives multiple > buildworlds just fine - so I'd say things are working pretty well. > > Steinar Haug, Nethelp consulting, sthaug@nethelp.no > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 14 14:14:58 2000 Delivered-To: freebsd-smp@freebsd.org Received: from milhouse.cnbinc.com (martin2.lnk.telstra.net [139.130.137.91]) by builder.freebsd.org (Postfix) with ESMTP id 0E2174FC7 for ; Mon, 14 Feb 2000 14:11:24 -0800 (PST) Received: from bender (bender.cnbinc.com [203.38.182.20]) by milhouse.cnbinc.com (8.9.3/8.9.3) with SMTP id IAA31168; Tue, 15 Feb 2000 08:38:50 +1030 (CST) (envelope-from diskiller@cnbinc.com) Message-ID: <001e01bf7738$0e9ff0a0$14b626cb@bender.cnbinc.com> From: "Martin Minkus" To: "Jordan K. Hubbard" , References: <48371.950560989@zippy.cdrom.com> Subject: Re: 4.0 SMP Problem? Date: Tue, 15 Feb 2000 08:38:40 +1030 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 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org ----- Original Message ----- From: Jordan K. Hubbard To: Cc: ; ; ; Sent: Tuesday, February 15, 2000 7:13 AM Subject: Re: 4.0 SMP Problem? > I have to echo these sentiments; I've been running softupdates on my > SMP box and building frequent releases on it, among other disk-intensive > exercises, and it works flawlessly. Has for months, in fact, despite > known-bugs getting fixed in soft updates during that time. > > - Jordan Hrmm .... interesting. (please read my other reply about the broken MP table). Perhaps there is something obscure in this box that FBSD 4.0 doesn't like too much. Perhaps the IDE controller, or the broken MP Table. I've only ONCE had a crash that gave me an intruction pointer (which was 0xc01dead8, which i found to be pretty funny (ie: dead :)). This was a custom kernel, and i did grep the symbols of /kernel but i forget what function it was in. Every other time, 80% of the time its a spontaneous reboot (doesn't say anything), the other 20% its a spontaneous crash. I must say, its very strange .... Whats REAL strange ... in the upgrade from 3.4 to 4.0, i built the kernel, rebooted, then the make buildworld, that all ran fine. Its as if make world (CPU, Disk acitivity) is indeed 100% fine. This machine's primary role is router, as well as sendmail, named, and apache. Even when 100% idle, and only just routing packets on the LAN, it crashed many time as 4.0. Perhaps its a bug with the ed0 driver, or tun0 driver ? (my guess is ed0 ?). I also have a WaveLAN in there, but i'm sure i tried it several times with the WaveLAN out, so i assume the wi0 driver is fine. So i only had tun0, and ed0, and once it dials up, it dies. I seriously had this machine idle 100% killing off pretty much every process, and just had it routing packets. 1 or 2 of us were just browsing the web, and, crash. It did it again ? I had a feeling it had something to do with network/routing .... I never tried changing the network card, however. They're all cheap PCI NE2000/Realtek clones. This is what *3.4* detects my card as... ed1: rev 0x00 int a irq 19 on pci0.17.0 ed1: address 00:40:05:3d:b1:85, type NE2000 (16 bit) 4.0 called it ed0. No, i don't have dmesg output. The entire machine was clobbered and reinstalled fresh as 3.4. Once i get some more spare hdd's i could have a play, i would love to have this resolved. martin. -- diskiller@cnbinc.com Linux/FreeBSD hacker Please use PGP: http://www.diskiller.net/pgp.html To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 14 14:15:49 2000 Delivered-To: freebsd-smp@freebsd.org Received: from milhouse.cnbinc.com (martin2.lnk.telstra.net [139.130.137.91]) by builder.freebsd.org (Postfix) with ESMTP id CE7B64CF3 for ; Mon, 14 Feb 2000 14:12:22 -0800 (PST) Received: from bender (bender.cnbinc.com [203.38.182.20]) by milhouse.cnbinc.com (8.9.3/8.9.3) with SMTP id IAA31174 for ; Tue, 15 Feb 2000 08:39:59 +1030 (CST) (envelope-from diskiller@cnbinc.com) Message-ID: <002d01bf7738$362c2580$14b626cb@bender.cnbinc.com> From: "Martin Minkus" To: Subject: Re: 4.0 SMP Problem? Date: Tue, 15 Feb 2000 08:39:49 +1030 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_002A_01BF7790.3898D7C0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org This is a multi-part message in MIME format. ------=_NextPart_000_002A_01BF7790.3898D7C0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable > On Mon, 14 Feb 2000, Martin Minkus wrote: > > [...]> > > SMP alone, and softupdate alone, and everything else, is broken on = 4.0. > > I cvsup'd a few days ago, so it was a recent current. > > I don't know about softupdates, which I do not use. SMP alone is *not* > broken on 4.0. It never was, not even during the recent (x)install > and lib problems. When i boot, it says "Broken MP Table Detected" Then it uses a work around. (dmesg output, 3.4-Stable) Intel Pentium detected, installing workaround for F00F bug APIC_IO: Testing 8254 interrupt delivery APIC_IO: Broken MP table detected: 8254 is not connected to IO APIC int = pin 2 APIC_IO: routing 8254 via 8259 on pin 0 changing root device to wd0s1a SMP: AP CPU #1 Launched! Perhaps that work around is broken on 4.0, which is what is causing me = the problems. I have NEVER had a problem with SMP on 3.x. I built a very simple 4.0 kernel with just the bare essentials. (ie, ad driver, and SMP) and it still panics and spontaneously reboots. -- diskiller@cnbinc.com Linux/FreeBSD hacker Please use PGP: http://www.diskiller.net/pgp.html ------=_NextPart_000_002A_01BF7790.3898D7C0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
> On Mon, 14 Feb 2000, Martin Minkus = wrote:
>
> [...]>
> > SMP alone, and softupdate = alone,=20 and everything else, is broken on 4.0.
> > I cvsup'd a few days = ago, so=20 it was a recent current.
>
> I don't know about softupdates, = which I=20 do not use. SMP alone is *not*
> broken on 4.0. It never was, not = even=20 during the recent (x)install
> and lib problems.

When i = boot, it=20 says "Broken MP Table Detected"
Then it uses a work = around.

(dmesg=20 output, 3.4-Stable)

Intel Pentium detected, installing workaround = for=20 F00F bug
APIC_IO: Testing 8254 interrupt delivery
APIC_IO: Broken = MP table=20 detected: 8254 is not connected to IO APIC int pin
2
APIC_IO: = routing 8254=20 via 8259 on pin 0
changing root device to wd0s1a
SMP: AP CPU #1=20 Launched!

Perhaps that work around is broken on 4.0, which is = what is=20 causing me the
problems. I have NEVER had a problem with SMP on = 3.x.

I=20 built a very simple 4.0 kernel with just the bare essentials. (ie, = ad
driver,=20 and SMP) and it still panics and spontaneously = reboots.

--
diskiller@cnbinc.com
Linux/Fr= eeBSD=20 hacker
 
Please use PGP: http://www.diskiller.net/pgp.h= tml
------=_NextPart_000_002A_01BF7790.3898D7C0-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 14 18: 2:26 2000 Delivered-To: freebsd-smp@freebsd.org Received: from smtp1.sisna.com (mail.sisna.net [209.210.176.10]) by builder.freebsd.org (Postfix) with ESMTP id 8A09A3E83 for ; Mon, 14 Feb 2000 18:02:24 -0800 (PST) Received: from mail.t-three.com [209.210.179.16] by smtp1.sisna.com with ESMTP (SMTPD32-6.00) id A35E956B010A; Mon, 14 Feb 2000 19:01:02 -0700 Received: from t-three.com [38.10.4.1] by mail.t-three.com with ESMTP (SMTPD32-5.05) id A35221EB0254; Mon, 14 Feb 2000 19:00:50 -0700 Message-ID: <38A8B3EA.C3C21932@t-three.com> Date: Mon, 14 Feb 2000 20:03:22 -0600 From: Jason Arnold Organization: Hypercaffeinated X-Mailer: Mozilla 4.7 [en] (X11; I; Linux 2.0.36 i386) X-Accept-Language: en MIME-Version: 1.0 To: freebsd-smp@freebsd.org Subject: Another compatible SMP motherboard for FreeBSD... Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I've just set up FreeBSD w/ SMP (2 x PII-450) on a Tyan 1836-DLUN. It works fine so if you want to add it to your web site, go ahead. --Jason Arnold To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 14 18: 4:16 2000 Delivered-To: freebsd-smp@freebsd.org Received: from synthcom.com (ip67.usw3.rb1.pdx.nwlink.com [207.202.141.67]) by builder.freebsd.org (Postfix) with ESMTP id 732A1404A for ; Mon, 14 Feb 2000 18:04:14 -0800 (PST) Received: from beacon.synthcom.com (beacon.synthcom.com [207.202.141.81]) by synthcom.com (8.9.3/8.9.3) with ESMTP id SAA54178 for ; Mon, 14 Feb 2000 18:18:36 -0800 (PST) (envelope-from neil@beacon.synthcom.com) Date: Mon, 14 Feb 2000 18:18:36 -0800 (PST) From: Neil Bradley Cc: freebsd-smp@freebsd.org Subject: Re: Another compatible SMP motherboard for FreeBSD... In-Reply-To: <38A8B3EA.C3C21932@t-three.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I've just set up FreeBSD w/ SMP (2 x PII-450) on a Tyan 1836-DLUN. It > works fine so if you want to add it to your web site, go ahead. Same deal with the Supermicro P6DNE. I'm running it with Dual PII-550's and 256MB of RAM. -->Neil ------------------------------------------------------------------------------- Neil Bradley Synthcom home : http://www.synthcom.com Synthcom Systems, Inc. ICQ # 29402898 "Hookt on Fonix" Workt four me! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 14 18:11:49 2000 Delivered-To: freebsd-smp@freebsd.org Received: from synthcom.com (ip67.usw3.rb1.pdx.nwlink.com [207.202.141.67]) by builder.freebsd.org (Postfix) with ESMTP id 4E7643DB9 for ; Mon, 14 Feb 2000 18:11:47 -0800 (PST) Received: from beacon.synthcom.com (beacon.synthcom.com [207.202.141.81]) by synthcom.com (8.9.3/8.9.3) with ESMTP id SAA54221 for ; Mon, 14 Feb 2000 18:26:09 -0800 (PST) (envelope-from neil@beacon.synthcom.com) Date: Mon, 14 Feb 2000 18:26:09 -0800 (PST) From: Neil Bradley To: freebsd-smp@freebsd.org Subject: Re: Another compatible SMP motherboard for FreeBSD... In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > I've just set up FreeBSD w/ SMP (2 x PII-450) on a Tyan 1836-DLUN. It > > works fine so if you want to add it to your web site, go ahead. > Same deal with the Supermicro P6DNE. I'm running it with Dual PII-550's > and 256MB of RAM. Whoops... I meant p6DBE. But my P6DNE (Dual Pentium Pro) is running this system on it just fine as well! -->Neil ------------------------------------------------------------------------------- Neil Bradley Synthcom home : http://www.synthcom.com Synthcom Systems, Inc. ICQ # 29402898 "Hookt on Fonix" Workt four me! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 14 18:13: 6 2000 Delivered-To: freebsd-smp@freebsd.org Received: from smtp1.sisna.com (mail.sisna.net [209.210.176.10]) by builder.freebsd.org (Postfix) with ESMTP id 8C3B53E5D for ; Mon, 14 Feb 2000 18:13:04 -0800 (PST) Received: from mail.t-three.com [209.210.179.16] by smtp1.sisna.com with ESMTP (SMTPD32-6.00) id A5DE95AD010A; Mon, 14 Feb 2000 19:11:42 -0700 Received: from t-three.com [38.10.4.1] by mail.t-three.com with ESMTP (SMTPD32-5.05) id A5D07B3400D8; Mon, 14 Feb 2000 19:11:31 -0700 Message-ID: <38A8B66B.8020235A@t-three.com> Date: Mon, 14 Feb 2000 20:14:03 -0600 From: Jason Arnold Organization: Hypercaffeinated X-Mailer: Mozilla 4.7 [en] (X11; I; Linux 2.0.36 i386) X-Accept-Language: en MIME-Version: 1.0 To: freebsd-smp@FreeBSD.ORG Subject: Re: Another compatible SMP motherboard for FreeBSD... References: <38A8B3EA.C3C21932@t-three.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Whoops, my bad for not reading the link. Thought that went to the maintainer of the SMP doc page... :( Jason Arnold wrote: > I've just set up FreeBSD w/ SMP (2 x PII-450) on a Tyan 1836-DLUN. It > works fine so if you want to add it to your web site, go ahead. > > --Jason Arnold > > 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 Tue Feb 15 3:31: 1 2000 Delivered-To: freebsd-smp@freebsd.org Received: from internal.mail.demon.net (internal.mail.demon.net [193.195.224.3]) by builder.freebsd.org (Postfix) with ESMTP id B1ACD3FC2; Tue, 15 Feb 2000 03:30:53 -0800 (PST) Received: from gti.noc.demon.net (gti.noc.demon.net [195.11.55.101]) by internal.mail.demon.net with ESMTP id LAA09920; Tue, 15 Feb 2000 11:31:12 GMT Received: from localhost (kevinw@localhost) by gti.noc.demon.net (8.8.8/8.8.8) with ESMTP id LAA00496; Tue, 15 Feb 2000 11:31:09 GMT X-Authentication-Warning: gti.noc.demon.net: kevinw owned process doing -bs Date: Tue, 15 Feb 2000 11:31:09 +0000 (GMT) From: Kevin Walton X-Sender: kevinw@gti.noc.demon.net To: Mike Smith Cc: freebsd-smp@freebsd.org Subject: Re: 8 way Xeon SMP on a IBM NetFinity 8500 In-Reply-To: <200002091641.IAA00710@mass.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 9 Feb 2000, Mike Smith wrote: > > We are getting an 8 way Xeon IBM NetFinity 8500 and want to test FreeBSD > > on it. Just wondering: > > > > 1) Does FreeBSD support 8 way Xeon SMP? > > 2) How 'well' does it support it? > > 3) Has anyone tried it on a NetFinity 8500, and did they have any > > problems? > > I expect that this will be a Corollary-based system, and probably not > Intel MP-spec compliant. If that's the case, we won't on it at all. I have spoken to IBM and have been told it is Intel MP-spec compliant, so we shall find out how it goes - ill let you know the results. Cheers Kev -- Kevin Walton Team Leader of Interactive Services Thus Data Services Demon Internet http://www.games.demon.net/ http://www.demon.net/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 15 7:54: 9 2000 Delivered-To: freebsd-smp@freebsd.org Received: from smtp01.primenet.com (smtp01.primenet.com [206.165.6.131]) by builder.freebsd.org (Postfix) with ESMTP id 0A5664BE4 for ; Tue, 15 Feb 2000 07:54:03 -0800 (PST) Received: (from daemon@localhost) by smtp01.primenet.com (8.9.3/8.9.3) id IAA14385; Tue, 15 Feb 2000 08:54:17 -0700 (MST) Received: from usr02.primenet.com(206.165.6.202) via SMTP by smtp01.primenet.com, id smtpdAAA9JaO6B; Tue Feb 15 08:54:06 2000 Received: (from tlambert@localhost) by usr02.primenet.com (8.8.5/8.8.5) id IAA18862; Tue, 15 Feb 2000 08:53:58 -0700 (MST) From: Terry Lambert Message-Id: <200002151553.IAA18862@usr02.primenet.com> Subject: Re: 4.0 SMP Problem? To: jrs@enteract.com (John Sconiers) Date: Tue, 15 Feb 2000 15:53:58 +0000 (GMT) Cc: diskiller@cnbinc.com (Martin Minkus), neil@synthcom.com (Neil Bradley), freebsd-smp@FreeBSD.ORG In-Reply-To: from "John Sconiers" at Feb 14, 2000 10:01:30 AM X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I'm using a pretty current SMP system and I'm not having these problems. > In my kernel I have but not using soffupdates. The only "big" problem I > saw was with sendmail core dumping. Is this 8.9.3, or 8.10beta12? Have you applied the patch for the deallocation of memory which occurs in a signal handler when child processes exit? Without this patch, it's fairly unstable under heavy load. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Feb 17 18:34:37 2000 Delivered-To: freebsd-smp@freebsd.org Received: from www.geocrawler.com (sourceforge.net [198.186.203.33]) by hub.freebsd.org (Postfix) with ESMTP id E2F8437B8EB for ; Thu, 17 Feb 2000 18:34:33 -0800 (PST) (envelope-from nobody@www.geocrawler.com) Received: (from nobody@localhost) by www.geocrawler.com (8.9.3/8.9.3) id SAA32078; Thu, 17 Feb 2000 18:34:37 -0800 Date: Thu, 17 Feb 2000 18:34:37 -0800 Message-Id: <200002180234.SAA32078@www.geocrawler.com> To: freebsd-smp@freebsd.org Subject: Intel L440GX supported From: "nefr0ma" Reply-To: "nefr0ma" Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org This message was sent from Geocrawler.com by "nefr0ma" Be sure to reply to that address. Hi, I just want to inform to people interesed that Intel L440GX server dual motherboad is *full* compatible with FreeBSD 3.4, I write this because, in the past, before buy it, noone was sure about its support. Well maybe someone is (or will be) in this situation I was; so, if you plan to buy it, got for it. dmesg: FreeBSD 3.4-RELEASE #4: Sat Dec 4 06:13:06 COT 2000 root@*********:/usr/src/sys/compile/************* Timecounter "i8254" frequency 1193182 Hz CPU: Pentium II/Xeon/Celeron (686-class CPU) Origin = "GenuineIntel" Id = 0x653 Stepping = 3 Features=0x183fbff real memory = 134217728 (131072K bytes) avail memory = 127713280 (124720K bytes) Programming 24 pins in IOAPIC #0 FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 1, version: 0x00040011, at 0xfee00000 cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 io0 (APIC): apic id: 2, version: 0x00170011, at 0xfec00000 Preloaded elf kernel "kernel" at 0xc02b6000. Pentium Pro MTRR support enabled Probing for devices on PCI bus 0: chip0: rev 0x00 on pci0.0.0 chip1: rev 0x00 on pci0.1.0 ahc0: rev 0x00 int a irq 19 on pci0.12. ahc0: aic7896/97 Wide Channel A, SCSI Id=7, 16/255 SCBs ahc1: rev 0x00 int a irq 19 on pci0.12. 1 ahc1: aic7896/97 Wide Channel B, SCSI Id=7, 16/255 SCBs fxp0: rev 0x08 int a irq 21 on pci0.14 .0 fxp0: Ethernet address 00:a0:c9:ac:a4:1e chip2: rev 0x02 on pci0.18.0 ide_pci0: rev 0x01 on pci0.18.1 chip3: rev 0x02 on pci0.18.3 vga0: rev 0x23 on pci0.2 0.0 Probing for devices on PCI bus 1: chip4: rev 0x06 on pci1.15.0 Probing for devices on PCI bus 2: Probing for PnP devices: Probing for devices on the ISA bus: sc0 on isa sc0: VGA color <16 virtual consoles, flags=0x0> atkbdc0 at 0x60-0x6f on motherboard atkbd0 irq 1 on isa sio0 at 0x3f8-0x3ff irq 4 flags 0x10 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: FIFO enabled, 8 bytes threshold fd0: 1.44MB 3.5in wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (wd0): wd0: 8063MB (16514064 sectors), 16383 cyls, 16 heads, 63 S/T, 512 B/S ppc0 at 0x378 irq 7 flags 0x40 on isa ppc0: Generic chipset (ECP/PS2/NIBBLE) in COMPATIBLE mode ppc0: FIFO with 16/16/8 bytes threshold lpt0: on ppbus 0 lpt0: Interrupt-driven port ppi0: on ppbus 0 vga0 at 0x3b0-0x3df maddr 0xa0000 msize 131072 on isa npx0 on motherboard npx0: INT 16 interface APIC_IO: Testing 8254 interrupt delivery APIC_IO: routing 8254 via pin 2 Waiting 15 seconds for SCSI devices to settle SMP: AP CPU #1 Launched! changing root device to da0s1a da0 at ahc1 bus 0 target 0 lun 0 da0: Fixed Direct Access SCSI-2 device da0: 40.000MB/s transfers (20.000MHz, offset 31, 16bit), Tagged Queueing Enabled da0: 4094MB (8386000 512 byte sectors: 255H 63S/T 522C) and the mptable: MPTable, version 2.0.15 MP Floating Pointer Structure: location: BIOS physical address: 0x000f6b70 signature: '_MP_' length: 16 bytes version: 1.4 checksum: 0x3e mode: Virtual Wire MP Config Table Header: physical address: 0x0009f960 signature: 'PCMP' base table length: 276 version: 1.4 checksum: 0xc5 OEM ID: 'INTEL ' Product ID: 'Lancewood ' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 26 local APIC address: 0xfee00000 extended table length: 164 extended table checksum: 225 MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 1 0x11 BSP, usable 6 5 3 0x183fb ff 0 0x11 AP, usable 6 5 3 0x183fb ff -- I/O APICs: APIC ID Version State Address 2 0x11 usable 0xfec00000 well, i think that's all, If someone search this mailing list with doubts about freebsd l440gx support, it's ok, it works ;) and when you read on the box "Design to maxime server uptime"...hrm, belive it ;) **********@***********$uptime 4:36PM up 87 days, 16 mins, 6 users, load averages: 1.31, 1.17, 1.17 p.s. Well, a bit more specific, all onboard chip (scsi, video, nic, ports,etc) are *full* compatible. Geocrawler.com - The Knowledge Archive To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message