Kernel panic on xen-unstable from 24.1.2005

Kernel panic on xen-unstable from 24.1.2005

[tdc]

I've tried unstable tarball in the hope it will solve my problems with 
non-working save / migrate on my machine, but using unstable branch, 
domU ends in kernel panic just from start.

Because errors occure in every Xen version i've tested so far, i suspect 
there is some hardware problem (maybe, but not probable, i had problems 
with save/migrate on all machines), or some incompatibility with certain 
gcc/glibc/whatever combination or maybe some obvious error in my 
dom0/domU kernel configuration (the same config worked - without 
save/migrate - on 2.0.3). Diff shows only slight changes between my 
config and the default one - cpu type, some device drivers and 
networking options.

Machine is built from common hardware - amd cpu, via kt600 chipset. 
There are 5 PCI add-on cards and only one empty PCI slot. Maybe there is 
IRQ routing problem?

If required, i can post additional data (output from ksymoops, both 
kernel, System.map files etc.)

Dave


---
last lines of 'xm dmesg':

(XEN) *** Serial input -> DOM0 (type 'CTRL-a' three times to switch 
input to Xen).
(XEN) PCI: Found IRQ 3 for device 00:0f.1
(XEN) IRQ routing conflict for 00:0f.1, have irq 14, want irq 3
(XEN) IRQ routing conflict for 00:12.0, have irq 4, want irq 3
(XEN) spurious 8259A interrupt: IRQ7.
(XEN) PCI: Found IRQ 11 for device 00:0e.0
(XEN) PCI: Found IRQ 12 for device 00:13.0
(XEN) PCI: Found IRQ 10 for device 00:0a.0
(XEN) PCI: Sharing IRQ 10 with 00:0d.0
(XEN) PCI: Sharing IRQ 10 with 01:00.0
(XEN) PCI: Found IRQ 3 for device 00:12.0
(XEN) IRQ routing conflict for 00:0f.1, have irq 14, want irq 3
(XEN) IRQ routing conflict for 00:12.0, have irq 4, want irq 3
(XEN) PCI: Found IRQ 9 for device 00:0c.0
(XEN) PCI: Found IRQ 10 for device 00:0d.0
(XEN) PCI: Sharing IRQ 10 with 00:0a.0
(XEN) PCI: Sharing IRQ 10 with 01:00.0

---
$ cat /proc/interrupts
            CPU0
   4:      49605        Phys-irq  eth0
   8:          2        Phys-irq  rtc
  10:       1047        Phys-irq  ide2
  11:        144        Phys-irq  libata
  12:        144        Phys-irq  libata
  14:      21944        Phys-irq  ide0
256:         20     Dynamic-irq  ctrl-if
257:     216739     Dynamic-irq  timer
258:          0     Dynamic-irq  console
259:          0     Dynamic-irq  net-be-dbg
NMI:          0
ERR:          0

---
startup output of domU with 48MB RAM:

Linux version 2.6.10-xenU (root@host) (gcc version 3.4.2 (Gentoo Linux 
3.4.2-r2, ssp-3.4.1-1, pie-8.7
.6.5)) #1 Tue Jan 25 21:00:47 CET 2005
BIOS-provided physical RAM map:
  Xen: 0000000000000000 - 0000000003000000 (usable)
48MB LOWMEM available.
DMI not present.
Built 1 zonelists
Kernel command line: 
ip=192.168.0.31:1.2.3.4:192.168.0.1.1:255.255.254.0:gentoo:eth0:off 
root=/dev/sda1 ro
Initializing CPU#0
PID hash table entries: 256 (order: 8, 4096 bytes)
Xen reported: 2000.103 MHz processor.
Using tsc for high-res timesource
Dentry cache hash table entries: 8192 (order: 3, 32768 bytes)
Inode-cache hash table entries: 4096 (order: 2, 16384 bytes)
Memory: 45764k/49152k available (1985k kernel code, 3324k reserved, 484k 
data, 132k init, 0k highmem)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
general protection fault: 0000 [#1]
PREEMPT
Modules linked in:
CPU:    0
EIP:    0061:[<c03753b0>]    Not tainted VLI
EFLAGS: 00010202   (2.6.10-xenU)
EIP is at init_amd+0x280/0x480
eax: 0383fbff   ebx: 00000006   ecx: c001001b   edx: 00000001
esi: c001001b   edi: c032a624   ebp: 00000000   esp: c036bf74
ds: 007b   es: 007b   ss: 0069
Process swapper (pid: 0, threadinfo=c036a000 task=c0327300)
Stack: c032a600 00000000 0000004a 00000000 c011aef8 00000030 00000400 
c02f9d18
        c036bfb8 c032a600 c03c3200 c032a600 c03c3200 c032a624 00000000 
c0374bee
        c032a600 0383fbff c1c3fbff 00000000 00000000 c036a000 c03c3200 
c0390aa0
Call Trace:
  [<c011aef8>] vprintk+0x168/0x220
  [<c0374bee>] identify_cpu+0x17e/0x1f0
  [<c036c18f>] check_bugs+0xf/0x60
  [<c036c7ef>] start_kernel+0x19f/0x200
  [<c036c330>] unknown_bootoption+0x0/0x1f0
Code: fa 3c 04 77 0e 8b 41 0c a9 00 00 00 02 0f 84 65 01 00 00 89 f1 80 
f9 08 0f 84 40 01 00 00 0f 86 48 01 00 00 be 1b 00 01 c0 89 f1 <0f> 32 
89 c7 25 00 00 f0 ff 89 54 24 18 3d 00 00 00 20 74 2c 89
  <0>Kernel panic - not syncing: Attempted to kill the idle task!
  <0>Rebooting in 1 seconds..





-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl

Re: Kernel panic on xen-unstable from 24.1.2005

[Keir Fraser]

The unstable tree is a non-starter for save/restore right now (or at
least, it definitely is for live migration). The SMP additions broke
certain other features which haven't yet been fixed again.

There are definitely save/restore bugs in 2.0, but we aim to fix those
soon. 

 -- Keir

> I've tried unstable tarball in the hope it will solve my problems with 
> non-working save / migrate on my machine, but using unstable branch, 
> domU ends in kernel panic just from start.
> 
> Because errors occure in every Xen version i've tested so far, i suspect 
> there is some hardware problem (maybe, but not probable, i had problems 
> with save/migrate on all machines), or some incompatibility with certain 
> gcc/glibc/whatever combination or maybe some obvious error in my 
> dom0/domU kernel configuration (the same config worked - without 
> save/migrate - on 2.0.3). Diff shows only slight changes between my 
> config and the default one - cpu type, some device drivers and 
> networking options.
> 
> Machine is built from common hardware - amd cpu, via kt600 chipset. 
> There are 5 PCI add-on cards and only one empty PCI slot. Maybe there is 
> IRQ routing problem?
> 
> If required, i can post additional data (output from ksymoops, both 
> kernel, System.map files etc.)
> 
> Dave


-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl

Re: Kernel panic on xen-unstable from 24.1.2005

[Ronald G. Minnich]

an old and common problem with K7 CPUs and vanilla kernels is that they'll
panic, since the default kernel tries to do things that the CPU won't do
(memory tells me they try to go after MSRs that are not there).  Redhat 8
installs always used to go fine on K7s but when you rebooted with the
installed kernel ... boom. The kernel they installed relied on features
not available on the k7. The symptom was always a GPF as you booted up.

This may not be your problem, but it looks strangely like that old problem 
which we used to see. You might want to comb over your kernel config 
again for your amd mobo.

ron



-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl