|
xen-devel
Re: [Xen-devel] Re: Xen-4.0.x doesn't start on 32bit / pvops 2.6.32.19
On 08/17/2010 07:16 AM, Pasi Kärkkäinen wrote:
> On Tue, Aug 17, 2010 at 01:04:38PM +0200, Stefan Kuhne wrote:
>> Am 16.08.2010 13:20, schrieb Pasi Kärkkäinen:
>>
>> Hello Pasi,
>>
>> Bootlog with your .config and actual Tree:
>> http://skweb.buetow.org/Linux/EisXen/xen.txt
>>
>>> And run:
>>> gdb vmlinux
>>>
>>> Then execute the following commands:
>>> (gdb) x/i 0xc17a56a1
>>> (gdb) list *0xc17a56a1
>>>
>>> (c17a56a1 is the EIP value from the crashlog).
>>>
>> (gdb) x/i 0xc08a2bf3
>> 0xc08a2bf3 <smp_scan_config+53>: cmpl $0x5f504d5f,(%ebx)
>> (gdb) list *0xc08a2bf3
>> 0xc08a2bf3 is in smp_scan_config (arch/x86/kernel/mpparse.c:700).
>> 695 bp, length);
>> 696 BUILD_BUG_ON(sizeof(*mpf) != 16);
>> 697
>> 698 while (length > 0) {
>> 699 mpf = (struct mpf_intel *)bp;
>> 700 if ((*bp == SMP_MAGIC_IDENT) &&
>> 701 (mpf->length == 1) &&
>> 702 !mpf_checksum((unsigned char *)bp, 16) &&
>> 703 ((mpf->specification == 1)
>> 704 || (mpf->specification == 4))) {
>> (gdb)
>>
>
> Jeremy: Does this look familiar? I don't think I have seen that..
Not specifically, but it might relate to a bug that Bastien reported
ages ago that I never followed up (attached).
J
--- Begin Message ---
On Mon, Oct 05, 2009 at 02:51:18PM -0700, Jeremy Fitzhardinge wrote:
> On 10/04/09 19:07, Bastian Blank wrote:
> > I'm not sure how to do it, but the paravirt-dom0 kernel needs PCBIOS PCI
> > probing under Xen disabled. It dies in nice tracebacks.
> Can you give an example?
| general protection fault: 0060 [#1] SMP
| last sysfs file:
| Modules linked in:
|
| Pid: 1, comm: swapper Tainted: G W (2.6.31-trunk-xen-686 #1)
| EIP: 0061:[<c00fb4bc>] EFLAGS: 00000046 CPU: 0
| EIP is at 0xc00fb4bc
| EAX: 12378000 EBX: 000f0000 ECX: 00000000 EDX: 0000b4c0
| ESI: c00f0280 EDI: c140cbb8 EBP: 00000000 ESP: dcc45f78
| DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: e021
| Process swapper (pid: 1, ti=dcc44000 task=dcc43960 task.ti=dcc44000)
| Stack:
| c144bc0d 00000060 00000001 00000001 dcc30220 c1006723 c144bf05 5f32335f
| <0> 3b1d067b 00000001 c134376b c144bab7 00000000 c144bae9 3b1d067b c146ae64
| <0> c1003084 3b1d067b c146ae64 c134376b 00000001 00000000 c141c476 3b1d067b
| Call Trace:
| [<c144bc0d>] ? pci_pcbios_init+0xe0/0x23d
| [<c1006723>] ? xen_restore_fl_direct_end+0x0/0x1
| [<c144bf05>] ? pci_direct_probe+0x95/0x1af
| [<c144bab7>] ? pci_arch_init+0x0/0x76
| [<c144bae9>] ? pci_arch_init+0x32/0x76
| [<c1003084>] ? do_one_initcall+0x63/0x175
| [<c141c476>] ? kernel_init+0x132/0x191
| [<c141c344>] ? kernel_init+0x0/0x191
| [<c100b267>] ? kernel_thread_helper+0x7/0x10
| Code: ef 66 ba fc 0c ed 3d 86 80 37 12 75 13 bb 00 00 0f 00 b9 00 00 00 00 ba
c0 b4 00 00 30 c0 eb 02 b0 80 81 64 24 08 fc ff ff ff 9d <cb> 00 00 00 9c fa 56
57 3c 01 75 14 66 bb 10 02 66 b9 00 00 ba
| EIP: [<c00fb4bc>] 0xc00fb4bc SS:ESP e021:dcc45f78
| ---[ end trace a7919e7f17c0a726 ]---
AFAIK this code just executes parts of the BIOS, which is not really
ment to work from ring 1.
> What's your kernel config?
It is a mostly standard Debian config, so it is large and you can find
it at http://hermes.jura.uni-tuebingen.de/~blank/debian/xen-test/config
Bastian
--
If I can have honesty, it's easier to overlook mistakes.
-- Kirk, "Space Seed", stardate 3141.9
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
--- End Message ---
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
|
|
Home