Subject: Re: Multiprocessor support
To: Manuel Bouyer <bouyer@antioche.eu.org>
From: Andreas Gustafsson <gson@gson.org>
List: port-xen
Date: 09/05/2007 20:05:33
Manuel Bouyer wrote:
> Maybe try without acpi ?
> No need to build a new kernel to try, just
> boot -c
> disable acpi
> quit
That's easier said than done, because when running under HVM, the
countdown after the "Press return to boot now, any other key for boot
menu" message goes by in the blink of an eye. Maybe qemu optimizes
away a delay loop or something?
Anyway, I built a new kernel:
include "arch/i386/conf/GENERIC.MP"
no fdc0 at isa?
no fd* at fdc?
no acpi0 at mainbus0
and tried it. It hung attaching the fourth CPU core:
>> NetBSD/i386 BIOS Boot, Revision 3.3
>> (root@guru.araneus.fi, Mon Aug 6 21:18:55 EEST 2007)
>> Memory: 639/261120 k
Press return to boot now, any other key for boot menu
booting hd0a:netbsd - starting in 0
8925552+407556+361624 [439424+422099]=0xa13594
BIOS CFG: Model-SubM-Rev: fc-00-01, 0x4074<EBDA,KBDINT,RTC,IC2,KBDF>
Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005,
2006, 2007
The NetBSD Foundation, Inc. All rights reserved.
Copyright (c) 1982, 1986, 1989, 1991, 1993
The Regents of the University of California. All rights reserved.
NetBSD 4.99.30 (NOACPI.MP) #0: Wed Sep 5 19:55:10 EEST 2007
root@dhcppc1:/usr/src/sys/arch/i386/compile/NOACPI.MP
total memory = 255 MB
rbus: rbus_min_start set to 0x40000000
avail memory = 240 MB
mainbus0 (root)
mainbus0: Intel MP Specification (Version 1.4) (_HVMCPU_ XEN )
cpu0 at mainbus0 apid 0: (boot processor)
cpu0: Intel Core 2 (Merom) (686-class), 2400.24 MHz, id 0x6f7
cpu0: "Intel(R) Core(TM)2 Quad CPU @ 2.40GHz"
cpu1 at mainbus0 apid 1: (application processor)
cpu1: Intel Core 2 (Merom) (686-class), 2400.69 MHz, id 0x6f7
cpu1: "Intel(R) Core(TM)2 Quad CPU @ 2.40GHz"
cpu2 at mainbus0 apid 2: (application processor)
cpu2: Intel Core 2 (Merom) (686-class), 2400.35 MHz, id 0x6f7
cpu2: "Intel(R) Core(TM)2 Quad CPU @ 2.40GHz"
cpu3 at mainbus0 apid 3: (application processor)
Again, when I tried to destroy the hung VM, the system locked up.
--
Andreas Gustafsson, gson@gson.org