Hi,
On Mon, Mar 12, 2012 at 1:54 PM, Konrad Rzeszutek Wilk
<konrad.wilk(a)oracle.com> wrote:
On Wed, Feb 29, 2012 at 11:04:34PM -0500, Sean McNamara wrote:
> Hi,
>
> My Fedora 17 3.3-rc4.git1 kernel boots on bare metal, but it dies very
> early on in the boot process when loading on top of Xen. It's got
> CONFIG_XEN_DOM0=y and all the other CONFIG_XEN* options are either "y"
> or "m". Linux panics with
>
> init[1] trap invalid opcode ip:7f99a072ff15 sp:7fffbf2a5088 error :0
> in libc-2.15.so[7f99a05f8000+1ac000]
> init used greatest stack depth: 2392 bytes left
> Kernel panic - not syncing: Attempted to kill init!
Hm, that is new. It looks like init dies and that stops the bootup process.
But I am not sure what would be happening to cause it.
Is this a 32-bit install or 64-bit?
64-bit.
I'll boot up my testing box on bare metal and re-test with the glibc
build from koji that purportedly fixes the issue. Thanks for your
help, Konrad and Major. I'll only be emailing back if I run into
further problems or if the issue is not fixed by RHBZ 801650
resolution.
Thanks,
Sean
>
> The kernel itself seems able to execute on top of Xen because I see a
> very normal-looking Xen init, then the screen blanks, then Linux
> initializes and starts executing early boot, stuff like the USB stack
> and SATA starts to come up, and then it tries to run the init process
> in userspace and it all comes crashing down.
>
> This is on Sandy Bridge hardware with all the CPU optional instruction
> sets enabled.
>
> Any ideas?
>
> Thanks,
>
> Sean
> --
> xen mailing list
> xen(a)lists.fedoraproject.org
>
https://admin.fedoraproject.org/mailman/listinfo/xen