Re: 2.6.15-rt16

[Clark Williams <williams@redhat.com>]

On Wed, 2006-02-01 at 20:22 -0500, Steven Rostedt wrote:
> Clark,
> 
> [...]
> switchroot: exec'ing /sbin/init
> init[1]: segfault at ffffffff8010fadc rip ffffffff8010fadc rsp 00007fffffdacfc8 error 15
> Kernel panic - not syncing: Segfault in init
> [...]
> 
> 
> Could you find where that ffffffff8010fadc is.  Compile with debug info,
> and (what I do) is a "gdb vmlinux" and "li *0xffffffff8010fadc" to find
> the location.  The dump may be something that is done after the init
> exits.

If the panic backtrace is to be believed, that's the address of mcount
(arch/x86_64/kernel/entry.S). 

The code 0x15 at the end of the segfault message means (according to the
comment in fault.c) that the segfault is the result of a user-mode
protection fault while reading the address ffffffff8010fadc. 

Clark

-- 
Clark Williams <williams@redhat.com>

Attachment: signature.asc
Description: This is a digitally signed message part