From owner-freebsd-smp Thu Feb 24 2:11:35 2000 Delivered-To: freebsd-smp@freebsd.org Received: from ipamzlx.physik.uni-mainz.de (ipamzlx.Physik.Uni-Mainz.DE [134.93.180.54]) by hub.freebsd.org (Postfix) with ESMTP id 4F94A37BBA8 for ; Thu, 24 Feb 2000 02:11:28 -0800 (PST) (envelope-from ohartman@ipamzlx.physik.uni-mainz.de) Received: from ipamzlx.Physik.Uni-Mainz.DE (ipamzlx.Physik.Uni-Mainz.DE [134.93.180.54]) by ipamzlx.physik.uni-mainz.de (8.9.3/8.9.3) with ESMTP id LAA27432 for ; Thu, 24 Feb 2000 11:11:39 +0100 (CET) (envelope-from ohartman@ipamzlx.physik.uni-mainz.de) Date: Thu, 24 Feb 2000 11:11:39 +0100 (CET) From: "O. Hartmann" To: freebsd-smp@freebsd.org Subject: MP Spec 1.1 vs 1.4 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ascii Content-Transfer-Encoding: 8BIT Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I have some confusing outputs of mptable. While reading some questions on Proliant systems from Compaq and others, I remember that my server reports to be MP Spec 1.1 (in the table got by mptable named Floating Point Structure ...). I think this is the MP Spec as discussed in manpages and in many postings. Our system is a custom made server, based on a dual board form GigaByte (GA6-BXD). I can switch by BIOS MP SPEC from 1.1 to 1.4 and I switched it to 1.4 without any affect getting another MP Spec report by mptable. What is wrong? I saw some other postings of users of 440NX chips based SMP systems and they got as mptable output Spec 1.4. I tried to install FBSD 4.0 on an older SMP system from Siemens Nixdorf. The machine is called PCE-5Smp, an EISA/ISA and PCI based Quad CPU system, here equippted with only two CPUs. This server uses a Mylex DAC960PD (it works fine with 4.0!!!) on PCI bus. Starting a single CPU kernel works, but sometimes I get some error messages about VM_XXX errors while copying a tar_ball file from a CDROM (attached to an aic0 controller, ISA Adaptec 1510). I try to get rid of this problem by getting the newest cvsupdated stuff and recompile the system. But, courios is: Windows NT Enterprise Server works fine on this older SMP system without any patches and fiddling around. Calling the NT own systutils for examine the system tells us having a SMP board with MP Spec 1.4. When calling mptable on an identical machine (they are really identical!) I get this output: =============================================================================== MPTable, version 2.0.15 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: EBDA physical address: 0x0009f0a0 signature: '_MP_' length: 16 bytes version: 1.1 checksum: 0x69 mode: PIC ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x0009fcb5 signature: 'ÿÿÿÿ' base table length: 65535 version: 1.255 checksum: 0xff OEM ID: 'ÿÿÿÿÿÿÿÿ' Product ID: 'ÿÿÿÿÿÿÿÿÿÿÿÿ' OEM table pointer: 0xffffffff OEM table size: 65535 entry count: 65535 local APIC address: 0xffffffff extended table length: 65535 extended table checksum: 255 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags -- Bus: Bus ID Type -- I/O APICs: APIC ID Version State Address -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ------------------------------------------------------------------------------- MP Config Extended Table Entries: Extended Table HOSED! That's it. Trying to start a SMP kerenl out of the blue fails, the system hangs. Has anybody any suggestions to these problems? Thanks in advance, Gruss O. Hartmann ------------------------------------------------------------------- ohartman@ipamzlx.physik.uni-mainz.de Klimadatenserver des IPA, Universitaet Mainz Netzwerk- und Systembetreuung To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message