Re: [Xen-devel] RE: [Xen-users] HVM WinXP dom crash (with 3.0.3 too)

[jdsw <jdsw2002@xxxxxxxxx>]

Hi
  With testing build and feedback from Keir to fix
hvmloader rebuilding, I was able to get HVM FC5
installed as guest domain. Felt really good to see HVM
working.

  After seeing the announcement, I downloaded the
3.0.3 PAE binaries and installed them. Created a
ramdisk and booted in to Xen 3.0.3 kernel. I am having
a bunch of issues... would REALLY appreciate if I can
resolve them  one by one. 

(Two critical of these are Network and  WinXP HVM
install, others are mentioned as them might be related
or genuine bugs)

1. I get the following message when I try to start
xend domain. (xend still seems to come up)

blktapctrl: error while loading shared libraries:
libcrypto.so.0.9.7: cannot open shared object file: No
such file or directory

-- These are crypto lib on my box --- seems to be
ahead!!!

/lib/libcrypto.so.0.9.8a
/lib/libcrypto.so.6
/usr/lib/libcrypto.a
/usr/lib/libcrypto.so
/usr/lib/pkgconfig/libcrypto.pc


2. When I boot, I get warning to move /lib/tls to
/lib/tls.disabled though I HAVE moved it to
/lib/tls.disabled.

3. It takes a minute or two before the login prompt
happens. No idea what the machine is doing during this
time.

4. My network is having following behaviour.
   4a : In dom0 ping works when ips are given, name
resolution does not work.
   4b : In DomU (FC5 HVM), dhcp does not work. Had to
manually config IP address, and name resolution does
not work.

bridge name     bridge id      STP enabled  
interfaces
xenbr0          8000.feffffffffff       no            
 peth0
                                                      
 vif0.0


5. WinXP installation : Windows setup screen comes up,
it shows a bunch of relavent drivers being loaded. It
just hangs for 10 min, then blue screen with 
STOP 0000000,....

xm dmesg shows :
===================
(XEN) (GUEST: 1) HVM Loader
(XEN) (GUEST: 1) Detected Xen v3.0.3-0
(XEN) (GUEST: 1) Writing SMBIOS tables ...
(XEN) (GUEST: 1) Loading ROMBIOS ...
(XEN) (GUEST: 1) Loading Cirrus VGABIOS ...
(XEN) (GUEST: 1) SVM go ...
(XEN) (GUEST: 1)  rombios.c,v 1.138 2005/05/07
15:55:26 vruppert Exp $
(XEN) (GUEST: 1) VGABios $Id: vgabios.c,v 1.61
2005/05/24 16:50:50 vruppert Exp $
(XEN) (GUEST: 1) HVMAssist BIOS, 1 cpu, $Revision:
1.138 $ $Date: 2005/05/07 15:55:26 $
(XEN) (GUEST: 1)
(XEN) (GUEST: 1) ata0-0: PCHS=16383/16/63
translation=lba LCHS=1024/255/63
(XEN) (GUEST: 1) ata0 master: QEMU HARDDISK ATA-7
Hard-Disk (10240 MBytes)
(XEN) (GUEST: 1) ata0  slave: Unknown device
(XEN) (GUEST: 1) ata1 master: QEMU CD-ROM ATAPI-4
CD-Rom/DVD-Rom
(XEN) (GUEST: 1) ata1  slave: Unknown device
(XEN) (GUEST: 1)
(XEN) (GUEST: 1) Booting from CD-Rom...
(XEN) (GUEST: 1) unsupported PCI BIOS function 0x0E
(XEN) (GUEST: 1) int13_harddisk: function 15, unmapped
device for ELDL=81
(XEN) (GUEST: 1) KBD: int09h_handler(): unknown
scancode read!

=================



--- jdsw <jdsw2002@xxxxxxxxx> wrote:

> Thanks Keir.
>    
>   My hvmloader was old, got the dev86 package from
> http://www.cix.co.uk/~mayday and rebuilt the things.
> Now I am passed that error. 
>    
>   btw, I had seen some traffic mentioning that there
> is no special need to build HVM stuff on one of the
> mailing list, which apparently is not true.
>    
>   My CDROM for windows is not bootable. Need to find
> the right CDROM and try it out.
>   (Is there a way to kick of a windows program like
> setup.exe...?)
>    
>   Still running in to most of other problems
> mentioned. I might repeat the whole procedure soon
> and report result to the forum.
>    
>   /Jd
>   
> 
> Keir Fraser <Keir.Fraser@xxxxxxxxxxxx> wrote:
>   If you?re still getting the bad hypercall message
> when the hvmloader runs, and thus unable to run
> ROMBIOS, then your hvmloader binary is still out of
> date. Where did you get 3.0.3-testing from? If you
> built yourself you need to download an extra 16-bit
> compiler (bcc). The instructions for this are listed
> when you try to build the HVM firmware (which fails
> to build if that compiler is absent). Look at
> /usr/lib/xen/boot/hvmloader ? what?s the timestamp
> on it?
> 
>  -- Keir
> 
> On 16/10/06 6:47 pm, "jdsw" <jdsw2002@xxxxxxxxx>
> wrote:
> 
>   Resending.... can anyone help?
> 
> Thanks
> /Jd
> 
> jdsw <jdsw2002@xxxxxxxxx> wrote:
>   Hi
>   I tried using xen-3.0.3-testing.hg  followed most
> of the stuff from 
> http://www.broermann.com/howto_amd_hvm.html
> Still getting the same problem when I try to start
> WinXP domU.  Same dmesg message about ROMBIOS. I
> would appreciate any help getting this and related
> issues resolved.
> 
> Q1  : Is the procedure ok (bit later in email) ? Do
> I need to do anything else ? (One interesting thing
> is that the kernel version that I had installed
> using yum, were 2.6.17-1.2187, while 3.0.3
> testing.hg is got 2.6.16.29 !!! )
> 
> Additionaly I am seeing the following. These are put
> here as they might be actual bugs or may help
> resolve the issue.
> 
> A . 
>  during startup, Between xend starting and login
> prompt takes 2 min.!!   I dont know whats going on ?
>  some messages in  dmesg 
> evbug.c: Event. Dev: isa0060/serio0/input0, Type: 4,
> Code: 4, Value: 50
> evbug.c: Event. Dev: isa0060/serio0/input0, Type: 1,
> Code: 50, Value: 1
> evbug.c: Event. Dev: isa0060/serio0/input0, Type: 0,
> Code: 0, Value: 0
> evbug.c: Event. Dev: isa0060/serio0/input0, Type: 4,
> Code: 4, Value: 18
> evbug.c: Event. Dev: isa0060/serio0/input0, Type: 1,
> Code: 18, Value: 1
> evbug.c: Event. Dev: isa0060/serio0/input0, Type: 0,
> Code: 0, Value: 0
> evbug.c: Event. Dev: isa0060/serio0/input0, Type: 4,
> Code: 4, Value: 50
> evbug.c: Event. Dev: isa0060/serio0/input0, Type: 1,
> Code: 50, Value: 0
> 
> B  
> -- Now my network name resolution is not working. I
> can ping using IP addresses.
> 
> C
> -- Xen is dropping 1GB memory of 4 GB!!! (see
> attached xm_dmesg)
>  Xen version 3.0.3-rc4  (root@) (gcc version 4.1.1
> 20060525 (Red Hat 4.1.1-1)) Thu Oct 12 22:58:53 PDT
> 2006
>  Latest ChangeSet: Thu Oct 12 19:09:08 2006 +0100
> 11758:946628333c31
> 
> (XEN) Command line: /xen.gz dom0_mem=768000
> (XEN) WARNING: Only the first 4 GB of the physical
> memory map can be accessed
> (XEN)          by Xen in 32-bit mode. Truncating the
> memory map...
> (XEN) Physical RAM map:
> (XEN)  0000000000000000 - 000000000009f800 (usable)
> (XEN)  000000000009f800 - 00000000000a0000
> (reserved)
> (XEN)  00000000000f0000 - 0000000000100000
> (reserved)
> (XEN)  0000000000100000 - 00000000bbff0000 (usable)
> (XEN)  00000000bbff0000 - 00000000bbff3000 (ACPI
> NVS)
> (XEN)  00000000bbff3000 - 00000000bc000000 (ACPI
> data)
> (XEN)  00000000f0000000 - 00000000f2000000
> (reserved)
> (XEN)  00000000fec00000 - 0000000100000000
> (reserved)
> (XEN) System RAM: 3007MB (3079740kB)
> (XEN) Xen heap:  10MB (10376kB)(XEN) PAE disabled.
> (XEN) found SMP MP-table at 000f5240
> (XEN) DMI 2.3 present.
> (XEN) Using APIC driver default
> (XEN) ACPI: RSDP (v000 GBT                          
>         ) @ 0x000f6bf0
> (XEN) ACPI: RSDT (v001 GBT    AWRDACPI 0x42302e31
> AWRD 0x01010101) @ 0xbbff3000
> (XEN) ACPI: FADT (v001 GBT    AWRDACPI 0x42302e31
> AWRD 0x01010101) @ 0xbbff3040
> (XEN) ACPI: SSDT (v001 PTLTD  POWERNOW 0x00000001 
> LTP 0x00000001) @ 0xbbff7a40
> (XEN) ACPI: MCFG (v001 GBT    AWRDACPI 0x42302e31
> AWRD 0x01010101) @ 0xbbff7c80
> (XEN) ACPI: MADT (v001 GBT    AWRDACPI 0x42302e31
> AWRD 0x01010101) @ 0xbbff79c0
> (XEN) ACPI: DSDT (v001 GBT    AWRDACPI 0x00001000
> MSFT 0x0100000c) @ 0x00000000
> (XEN) ACPI: Local  APIC address 0xfee00000
> (XEN) ACPI: LAPIC (acpi_id[0x00] lapic_id[0x00]
> enabled)
> (XEN) Processor #0 15:11 APIC version 16
> 
> 
> ============  BUILD PROCEDURE ===================
> # cd /usr/src/
> # hg clone
> http://xenbits.xensource.com/xen-3.0.3-testing.hg
> # cd xen-3.0-3testing.hg
> # make world
> # make install
> # mkinitrd initrd.img-2.6.16.29-xen 2.6.16.29-xen
> # ln -sf initrd.img-2.6.16.29-xen initrd.img-2.6-xen
> 
> --- disabled ASUS-ACPI
> # make linux-2.6-xen-config CONFIGMODE=menuconfig
> # make linux-2.6-xen-build
> # make linux-2.6-xen-install
> 
> -- did the whole procedure twice.. once skipped this
> two
> --- steps (removed and rebuilt, as mkinitrd was
>  compalaining..
> # rm initrd.img-2.6.16.29-xen 
> # mkinitrd initrd.img-2.6.16.29-xen  2.6.16.13-xen
> 
> I had to manually copy xen and grub from
> /usr/lib/python to /usr/lib/python2.4/site-packages.
> Added a new entry in the menu.lst under
>  grub.
> 
> title Fedora Core TEST (2.6.16.29-xen)
>         root (hd0,0)
>         kernel /xen.gz dom0_mem=768000
>         module /vmlinuz-2.6.16.29-xen ro
> root=/dev/VG_Fedora/LV2 rhgb quiet
>         module /initrd.img-2.6.16.29-xen
> 
> booted in Test kernel. And tried to create HVM winxp
> domu.
> ===============
> 
> 
> Keir Fraser <Keir.Fraser@xxxxxxxxxxxx> wrote:
>   Re: [Xen-devel] RE: [Xen-users] HVM WinXP dom
> crash     On 9/10/06 8:19 pm, "jdsw"
> <jdsw2002@xxxxxxxxx> wrote:
>  
>  
>   ok.. I would not worry about the WRMSR warnings.
>  But how do I get pass the ROMBIOS problem ?
>  
>  ccing devloper forum,in case someone knows about
> this  particular issue.
>  
> 
>  The hvmloader binary is definitely out of sync with
> the Xen binary (hvmloader?s SVM hypercall interface
> is older than the Xen that you are running knows
> about). This might be a misfeature of the FC5
> packages, although I haven?t heard anyone else
> report this problem.
> 
=== message truncated ===


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

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