Re: -git: BUG: unable to handle kernel paging request at c0100248 [CPU hotplug]

["Vegard Nossum" <vegard.nossum@gmail.com>]

On Sun, Jul 27, 2008 at 11:55 AM, Vegard Nossum <vegard.nossum@gmail.com> wrote:
> Hi,
>
> Something broke CPU-hotplug _again_, here with a freshly built latest -git:
>
> BUG: unable to handle kernel paging request at c0100248

[snip]

> c067c3db:       c7 05 48 02 10 c0 f0    movl   $0xc067c9f0,0xc0100248
> c067c3e2:       c9 67 c0
>
> $ addr2line -e vmlinux -i c067c3db
> arch/x86/kernel/smpboot.c:859
> arch/x86/kernel/smpboot.c:993
>
> Hm... that doesn't really make sense, does it? It's this line:
>
>        initial_code = (unsigned long)start_secondary;
>
> ..but I don't see where the memory dereference comes from.

"initial_code" (0xc0100248) is in a .text section:

Idx Name          Size      VMA       LMA       File off  Algn
  0 .text.head    000003a1  c0100000  00100000  00001000  2**2
  1 .text         00583b8d  c01003b0  001003b0  000013b0  2**4

And I have CONFIG_DEBUG_RODATA=y. Maybe this was introduced in

commit e3f77edfc1d0beb7b10f9f31d9e39206f7dbef7b
Author: Glauber Costa <gcosta@redhat.com>
Date:   Wed May 28 12:57:02 2008 -0300

    x86: use initial_code for i386

    x86_64 jumps to whatever is written in "initial_code" symbol,
    instead of a fixed address. Do it for i386 too. It will allow us
    to integrate more of the smp boot code.

    Signed-off-by: Glauber Costa <gcosta@redhat.com>
    Signed-off-by: Ingo Molnar <mingo@elte.hu>

?


Vegard

-- 
"The animistic metaphor of the bug that maliciously sneaked in while
the programmer was not looking is intellectually dishonest as it
disguises that the error is the programmer's own creation."
	-- E. W. Dijkstra, EWD1036