[Xen-users] About kernel smp and acpi on hardware vm.

["Jean-Michel Bonnefond" <pompon2@xxxxxxxxx>]

Hello list,

I have a problem with my new xen server wich can't reach to start my linux hardware virtual machines which works on my old server without modifications.

Here is my situation :
On all xen servers (3), I use xen 3.2 for a Dom0 compiled with the linux kernel caming from http://xenbits.xensource.com/linux-2.6.18-xen.hg

My 2 old xen servers are :
Dual core Intel(R) Pentium(R) D CPU 2.80GHz
Ubuntu 7.10
#uname -a : Linux 2.6.18.8-xen #1 SMP Thu Jan 24 17:04:50 CET 2008 i686 GNU/Linux
#getconf WORD_BIT : 32

My new xen server is :
Quadri core Intel(R) Xeon(R) CPU E5335  @ 2.00GHz
Ubuntu 7.10
#uname -a : Linux 2.6.18.8-xen #1 SMP Thu Mar 6 16:32:32 CET 2008 i686 GNU/Linux
#getconf WORD_BIT : 32

My hvm server is :
Disk is emulated in a 3Gb file (ex : disk = [ 'file:/xen/hvm-mic.dsk,hda,w' ]
Debian etch
#uname -a : Linux 2.6.18-4-686 #1 SMP Wed Feb 21 16:06:54 UTC 2007 i686 GNU/Linux
#getconf WORD_BIT : 32


My hvm server starts without problem on the old xen server, but when I try to start it on the new xen server, It freeze after this message : 
[...]
>CPU: L1 I cache: 32K, L1 D cache: 32K
>CPU: L2 cache: 4096K
>Intel machine check architecture supported
>Intel machine check reporting enabled on CPU#0
>CompatvDSO mapped to ffffe000.
>Checking 'htl' instruction... OK.
>SMP alternatives: switching to UP code
>Freeing SMP alternatives: 16k freed  
>ACPI: Core revision 20060707
>CPU0: Intel(R) Xeon(R) CPU    E5335  @ 2.00GHz stepping 0b
>weird, boot CPU (#0) not listed by the BIOS.


if I add the 'nosmp' option to the linux kernel of the hvm machine, the boot process does advance a bit more, but cycle on these messages :
[...]
>ACPI Exception (acpi_processor-0681): AE_NOT_FOUND, Processor Device is not present [20060707]
>ACPI: Getting cpuindex for acpiid 0x1
>ACPI Exception (acpi_processor-0681): AE_NOT_FOUND, Processor Device is not present [20060707]
>ACPI: Getting cpuindex for acpiid 0x2
>ACPI Exception (acpi_processor-0681): AE_NOT_FOUND, Processor Device is not present [20060707]
>ACPI: Getting cpuindex for acpiid 0x3
>Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
>ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
>PIIX3: IDE IDE controller at PCI slot 0000:00:01.1
>PIIX3: chipset revision 0
>PIIX3: not 100% bative mode: will probe irqs later
>   ide0: BM-DMA at 0xc000-0xc007, BIOS settings: hda:pio, hdb:pio
>   ide1: BM-DMA at 0xc000-0xc00f, BIOS settings: hdc:pio, hdd:pio
>8139cp: 10/100 PCI Ethernet driver v1.2 (Mar 22, 2004)
>hda: QEMU HARDDISK, ATA DISK drive
>ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
>ACPI: PCI Interrupt 0000:00:04.0[A] -> GSI 32 (level, low) -> IRQ 57
>eth0: RTL-8139C+ at 0xc8814000, 00:16:3e:71:56:22, IRQ 57
>8139too Fast Ethernet driver 0.9.27
>hda: max request size: 512KiB
>hda: 6144000 sectors (3145 MB) w/256KiB Cache, CHS=6095/255/63, (U)DMA
>hda: lost interrupt
>hda: lost interrupt
>hda: lost interrupt
>hda: set_multmode: status=0x41 { DriveReady Error }
>hda: set_multmode: error=0x04 { DriveStatusError}
>ide: failed opceode was: 0xef
>hda: lost interrupt
>hda: lost interrupt
>hda: cache flushes supported
> hda:<4>hda: dma_timer_expiry: dma status == 0x24
>hda: DMA interript recovery
>hda: lost interrupt
> hda1 hda2 hda3
>Done.
>Begin: Mounting root file system... ...
>Begin: Running /scripts/local-top ...
>Done.
>Begin: Waiting for root file system... ...
>hda: dma_timer_expiry: dma status == 0x24
>hda: DMA interript recovery
>hda: lost interrupt
>hda: lost interrupt
>hda: lost interrupt
>hda: dma_timer_expiry: dma status == 0x24
>hda: DMA interript recovery
>hda: lost interrupt
>hda: dma_timer_expiry: dma status == 0x24
>hda: DMA interript recovery
>hda: lost interrupt
>hda: dma_timer_expiry: dma status == 0x24
>hda: DMA interript recovery
>hda: lost interrupt



Now if I add 'acpi=off' and 'nosmp' option to the linux kernel of the hvm machine, the virtual machine finaly reach to boot.

This works but it is not a very good solution for me as I can't automatically move forth and back my hvms from old to new xen server. Also the acpi=off don't stop the machine on the xen side when halting.

Do you have any idea of what could be the root cause of this behavior? 
Of course I could provide any supplementary information I should have forgot to give in this email.

Thanks you,
Jean-Michel.

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users