Full virtualization with Pacifica crashes

Full virtualization with Pacifica crashes

[Alexander Osthof]

[-- Attachment #1.1.1: Type: text/plain, Size: 1197 bytes --]

I'm currently trying to get Xen working on an Pacifica system. But each time I 
try to start a fully virtualized guest domain, the system hangs just after 
the start with a black screen (paravirtualization works, btw).

I've also tried to monitor the system with the serial console, but I get 
nearly no useful info. If I don't specify the memory used by dom0 (with boot 
option "dom0_mem=xxx"), I can see the following error:

  Unable to open display.
  (XEN) domain_crash called from shadow_public.c:1564
  (XEN) Domain 1 reported crashed by domain 0 on cpu#0:

If I specify explicitely the amount of memory to be used by dom0, no error 
message appears, but the screen wents black, too. So I don't really know if 
to blame shadow_public.c and thus memory mapping respectively.

Have you ever been successful in running such a full virtualized guest domain 
on a Pacifica system, or do you have a clue or a secret tip how to manage to 
get this done?

Attached is the config file for the guest domain.

-- 
Alexander Osthof				R&D SI Architecture Maintenance
SUSE Linux Products GmbH, Maxfeldstr.5		Phone: +49-911-74053-285
D-90409 Nuernberg				EMail: aosthof@suse.de

[-- Attachment #1.1.2: PacificaDomU --]
[-- Type: text/plain, Size: 5801 bytes --]

#  -*- mode: python; -*-
#============================================================================
# Python configuration setup for 'xm create'.
# This script sets the parameters used when a domain is created using 'xm create'.
# You use a separate script for each domain you want to create, or 
# you can set the parameters for the domain on the xm command line.
#============================================================================

import os, re
arch = os.uname()[4]
if re.search('64', arch):
    arch_libdir = 'lib64'
else:
    arch_libdir = 'lib'

#----------------------------------------------------------------------------
# Kernel image file.
kernel = "/usr/lib/xen/boot/hvmloader"

# The domain build function. HVM domain uses 'hvm'.
builder='hvm'

# Initial memory allocation (in megabytes) for the new domain.
#memory = 128
memory = 512

# A name for your domain. All domains must have different names.
name = "MySusePacifica"

#-----------------------------------------------------------------------------
# the number of cpus guest platform has, default=1
#vcpus=1

# enable/disable HVM guest PAE, default=0 (disabled)
#pae=0

# enable/disable HVM guest ACPI, default=0 (disabled)
#acpi=0

# enable/disable HVM guest APIC, default=0 (disabled)
#apic=0

# List of which CPUS this domain is allowed to use, default Xen picks
#cpus = ""         # leave to Xen to pick
#cpus = "0"        # all vcpus run on CPU0
#cpus = "0-3,5,^1" # run on cpus 0,2,3,5

# Optionally define mac and/or bridge for the network interfaces.
# Random MACs are assigned if not given.
#vif = [ 'type=ioemu, mac=00:16:3e:00:00:11, bridge=xenbr0' ]
# type=ioemu specify the NIC is an ioemu device not netfront
vif = [ 'type=ioemu, bridge=xenbr0' ]

#----------------------------------------------------------------------------
# Define the disk devices you want the domain to have access to, and
# what you want them accessible as.
# Each disk entry is of the form phy:UNAME,DEV,MODE
# where UNAME is the device, DEV is the device name the domain will see,
# and MODE is r for read-only, w for read-write.

#disk = [ 'phy:hda1,hda1,r' ]
#disk = [ 'file:/var/lib/xen/images/disk.img,ioemu:hda,w' ]
disk = [ 'phy:sda7,hda1,w' ]

#----------------------------------------------------------------------------
# Configure the behaviour when a domain exits.  There are three 'reasons'
# for a domain to stop: poweroff, reboot, and crash.  For each of these you
# may specify:
#
#   "destroy",        meaning that the domain is cleaned up as normal;
#   "restart",        meaning that a new domain is started in place of the old
#                     one;
#   "preserve",       meaning that no clean-up is done until the domain is
#                     manually destroyed (using xm destroy, for example); or
#   "rename-restart", meaning that the old domain is not cleaned up, but is
#                     renamed and a new domain started in its place.
#
# The default is
#
#   on_poweroff = 'destroy'
#   on_reboot   = 'restart'
#   on_crash    = 'restart'
#
# For backwards compatibility we also support the deprecated option restart
#
# restart = 'onreboot' means on_poweroff = 'destroy'
#                            on_reboot   = 'restart'
#                            on_crash    = 'destroy'
#
# restart = 'always'   means on_poweroff = 'restart'
#                            on_reboot   = 'restart'
#                            on_crash    = 'restart'
#
# restart = 'never'    means on_poweroff = 'destroy'
#                            on_reboot   = 'destroy'
#                            on_crash    = 'destroy'

#on_poweroff = 'destroy'
#on_reboot   = 'restart'
#on_crash    = 'restart'

#============================================================================

# New stuff
device_model = '/usr/' + arch_libdir + '/xen/bin/qemu-dm'

#-----------------------------------------------------------------------------
# Disk image for 
#cdrom=

#-----------------------------------------------------------------------------
# boot on floppy (a), hard disk (c) or CD-ROM (d) 
#boot='d'
#-----------------------------------------------------------------------------
#  write to temporary files instead of disk image files
#snapshot=1

#----------------------------------------------------------------------------
# enable SDL library for graphics, default = 0
sdl=0

#----------------------------------------------------------------------------
# enable VNC library for graphics, default = 1
vnc=1

#----------------------------------------------------------------------------
# enable spawning vncviewer(only valid when vnc=1), default = 1
vncviewer=1

#----------------------------------------------------------------------------
# no graphics, use serial port
#nographic=0

#----------------------------------------------------------------------------
# enable stdvga, default = 0 (use cirrus logic device model)
stdvga=0

#-----------------------------------------------------------------------------
#   serial port re-direct to pty deivce, /dev/pts/n 
#   then xm console or minicom can connect
serial='pty'

#----------------------------------------------------------------------------
# enable ne2000, default = 0(use pcnet)
ne2000=0


#-----------------------------------------------------------------------------
#   enable audio support
#audio=1


#-----------------------------------------------------------------------------
#    set the real time clock to local time [default=0 i.e. set to utc]
#localtime=1


#-----------------------------------------------------------------------------
#    start in full screen
#full-screen=1   

[-- Attachment #1.2: Type: application/pgp-signature, Size: 189 bytes --]

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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

Re: Full virtualization with Pacifica crashes

[Tristan Gingold]

Le Mardi 25 Avril 2006 15:20, Alexander Osthof a écrit :
> I'm currently trying to get Xen working on an Pacifica system. But each
> time I try to start a fully virtualized guest domain, the system hangs just
> after the start with a black screen (paravirtualization works, btw).
>
> I've also tried to monitor the system with the serial console, but I get
> nearly no useful info. If I don't specify the memory used by dom0 (with
> boot option "dom0_mem=xxx"), I can see the following error:
>
>   Unable to open display.
>   (XEN) domain_crash called from shadow_public.c:1564
>   (XEN) Domain 1 reported crashed by domain 0 on cpu#0:
>
> If I specify explicitely the amount of memory to be used by dom0, no error
> message appears, but the screen wents black, too. So I don't really know if
> to blame shadow_public.c and thus memory mapping respectively.
>
> Have you ever been successful in running such a full virtualized guest
> domain on a Pacifica system, or do you have a clue or a secret tip how to
> manage to get this done?
>
> Attached is the config file for the guest domain.
Your disk= configuration line seems a little bit strange: you are using a 
non-hvm configuration.

At this time, hvm only accepts a file as a disk image.  You can't use a 
partition nor a disk!

Tristan.

Re: Full virtualization with Pacifica crashes

[Andi Kleen]

Alexander Osthof <aosthof@suse.de> writes:

> I'm currently trying to get Xen working on an Pacifica system. But each time I 
> try to start a fully virtualized guest domain, the system hangs just after 
> the start with a black screen (paravirtualization works, btw).
> 
> I've also tried to monitor the system with the serial console, but I get 
> nearly no useful info. If I don't specify the memory used by dom0 (with boot 
> option "dom0_mem=xxx"), I can see the following error:
> 
>   Unable to open display.

This means your $DISPLAY is wrong and it can't open the window
and it probably goes downwards after that. The problem is that xend
not xm tries to open the window and it not always gets all the environment
variables needed for it. There were some fixes in this area but
maybe they were not enough. 

One workaround is to restart xend on your current terminal to fix
up its environment.

Of course even with that it shouldn't hang, but error handling paths
tend to be often undertested.

-Andi

RE: Full virtualization with Pacifica crashes

[Ian Pratt]

 
> Your disk= configuration line seems a little bit strange: you 
> are using a non-hvm configuration.
> 
> At this time, hvm only accepts a file as a disk image.  You 
> can't use a partition nor a disk!

That's not true at all. Both LVM and raw partitions work fine as well as
files.

Ian

Re: Full virtualization with Pacifica crashes

[Tristan Gingold]

Le Mardi 25 Avril 2006 15:41, Ian Pratt a écrit :
> > Your disk= configuration line seems a little bit strange: you
> > are using a non-hvm configuration.
> >
> > At this time, hvm only accepts a file as a disk image.  You
> > can't use a partition nor a disk!
>
> That's not true at all. Both LVM and raw partitions work fine as well as
> files.
Humm, maybe I spoke to fast.  I didn't check recently, but about 1 month ago, 
I had such a problem with xen/ia64+VTi.

Tristan.

Re: Full virtualization with Pacifica crashes

[Ray Bryant]

On Tuesday 25 April 2006 08:41, Ian Pratt wrote:
> > Your disk= configuration line seems a little bit strange: you
> > are using a non-hvm configuration.
> >
> > At this time, hvm only accepts a file as a disk image.  You
> > can't use a partition nor a disk!
>
> That's not true at all. Both LVM and raw partitions work fine as well as
> files.
>

How does one set that up in the config file?    In particular, does one 
specify the device as a phys or an ioemu device?
> Ian
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel

-- 
Ray Bryant
AMD Performance Labs                   Austin, Tx
512-602-0038 (o)                 512-507-7807 (c)

RE: Full virtualization with Pacifica crashes

[Ian Pratt]

> > > Your disk= configuration line seems a little bit strange: you are 
> > > using a non-hvm configuration.
> > >
> > > At this time, hvm only accepts a file as a disk image.  You can't 
> > > use a partition nor a disk!
> >
> > That's not true at all. Both LVM and raw partitions work 
> fine as well 
> > as files.
> >
> 
> How does one set that up in the config file?    In 
> particular, does one 
> specify the device as a phys or an ioemu device?

You specify it exactly the same was as you do for pv devices, just
putting the ioemu: prefix on the 2nd (guest target device) stanza. e.g.

  disk = [ 'phy:/dev/vg/my_disk, ioemu:hda, w' ]

Ian

RE: Full virtualization with Pacifica crashes

[Woller, Thomas]

> 
> Attached is the config file for the guest domain.
Specify cpus= explicitly for core 0 (cpus = "0") for your guest, we are
having issues with using a Dom0 SMP with 64bit hv, and then attempting
to start/pin unmodified guests to cores>0 (system reboots).  32bit hv
works fine with SMP Dom0/hv on cores>0.

Also, try using recommendation by Ian for phys
>   disk = [ 'phy:/dev/vg/my_disk, ioemu:hda, w' ]

I've installed win2003 server from cdrom=iso to physical disk /dev/hdb
(100Gig partition) with this method, so it does work for SVM, using
core0 for guest.
disk=['phy:/dev/hdb,ioemu:hda,w']
Specifically.

Can you post output from /proc/cpuinfo from Dom0?

Can you provide last line on serial output where the system hangs when
starting up an unmodified guest? (if no valid output, that's fine the
cpuinfo will help).

thanks
tom

Re: Full virtualization with Pacifica crashes

[Alexander Osthof]

[-- Attachment #1.1: Type: text/plain, Size: 1130 bytes --]

Am Donnerstag, 27. April 2006 18:41 schrieb Woller, Thomas:
> > Attached is the config file for the guest domain.
>
> Specify cpus= explicitly for core 0 (cpus = "0") for your guest, we are
> having issues with using a Dom0 SMP with 64bit hv, and then attempting
> to start/pin unmodified guests to cores>0 (system reboots).  32bit hv
> works fine with SMP Dom0/hv on cores>0.
>
Thanks a lot, that solved the problem. I'm able now to install unmodified 
guests.

> Also, try using recommendation by Ian for phys
>
> >   disk = [ 'phy:/dev/vg/my_disk, ioemu:hda, w' ]
>
> I've installed win2003 server from cdrom=iso to physical disk /dev/hdb
> (100Gig partition) with this method, so it does work for SVM, using
> core0 for guest.
> disk=['phy:/dev/hdb,ioemu:hda,w']
> Specifically.
>
> Can you post output from /proc/cpuinfo from Dom0?
>
I think you don't need the output anymore, do you?

> Can you provide last line on serial output where the system hangs when
> starting up an unmodified guest? (if no valid output, that's fine the
> cpuinfo will help).
>
> thanks
> tom

Kind regards, 
Alex

[-- Attachment #1.2: Type: application/pgp-signature, Size: 189 bytes --]

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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