kernel 2.6.27.5-37 doesn't boot after recompile in F9

[stan]

stan wrote:
> Hi,
> 
> This is a longshot but I've been downloading the source for the fedora 
> kernels and compiling them to suit my system better for 6 months or so 
> (since 2.6.24).  It has been working great.  I've been using the same 
> config since I got one working (there seem to be all kinds of 
> undocumented dependencies between options in the kernel, so a little 
> trial and error is necessary to get one that flies).  I just run make 
> oldconfig and update the .config and away it goes.
> I don't modify any of the setup, so this is the Fedora kernel recompiled 
> with all of its patches.
> 
> Well, it seemed to work with 2.6.27 as it compiled cleanly.  But after I 
> installed it and built the initrd, it won't boot.  The message is about 
> a segmentation fault, error 4. It complains that init can't start and 
> that the interrupts are not synching and hangs.  I've tried many options 
> but the error is always the same.
> 
> I installed the official 2.6.27 rpm package for f9 and it boots and runs 
> just fine.
> 
> Has anyone else seen this, and did you get it to work somehow?  I am 
> suspicious of the mkinitrd as I don't think it updated with the update 
> to 2.6.27, but I don't know.
> 
> Thanks for any insight.
> 
Well, after much plodding, it seems that there is a warning 
that is given during compile if the right switch is turned 
on.  make CONFIG_DEBUG_SECTION_MISMATCH=y

Here is that warning:

CC      arch/x86/lib/usercopy_32.o
AR      arch/x86/lib/lib.a
LD      vmlinux.o
MODPOST vmlinux.o
WARNING: vmlinux.o(.cpuinit.data+0x0): Section mismatch in 
reference from the variable initial_code to the function
.init.text:i386_start_kernel()
The variable __cpuinitdata initial_code references
a function __init i386_start_kernel().
If i386_start_kernel is only used by initial_code then
annotate i386_start_kernel with a matching annotation.

GEN     .version
CHK     include/linux/compile.h
UPD     include/linux/compile.h


Here is the error when that kernel was booted after 
installation.

init[1]: segfault at aa4830 ip 00aa4830 sp bfea2390 error 4
Kernel Panic - not syncing: Aiee, killing interrupt handler!

It prints this message just before the penguin and nash 
message come up in the boot sequence.  Seems to be a time 
when the kernel would be starting.

I don't know if it is something I'm doing in the 
configuration or if the mismatch on the kernel init is 
related.  I know that Fedora's policy is if you break it, 
you've bought it, so I'm not looking for any resolution. 
I'm just putting it out here for someone to find if they 
have the same problem.  I've made a lot of changes in the 
configuration, so that could be it.  Not sure.