2.5.68-mm2+e100=trouble

Hello -

This may be of interest...

kernel: 2.5.68-mm2

Linux distro: Red Hat 8.0 + updates

Hardware:
Celeron 1.2 Ghz on Intel Motherboard
512 MB RAM, 2x e100 ethernet

2.5.68-mm2 initially would not boot for me, and I noticed that it was dying at the point where it was loading the e100 driver -

I booted -mm2 again specifying the eepro100 driver instead and it came up fine.

To get another look at the e100 problem, I shut down networking and removed the eepro100 module - all good, so far -

Then I said "modprobe e100" and the following oops occurred:

-------------------------- snip ------------------------------

Unable to handle kernel paging request at virtual address c048f190

printing eip:
c048f190
*pde = 00103027
*pte = 0048f000

Oops: 0000 [#1]
CPU:    0
EIP:    0060:[<c048f190>]    Not tainted VLI
EFLAGS: 00010286

EIP is at apply_alternatives+0x0/0xf0
eax: e0e215be ebx: e0e0f618 ecx: 0000008b edx: 00000076
esi: c03df35b edi: e0e0f36e ebp: d566defc esp: d566dee0
ds: 007b es: 007b ss: 0068
Process modprobe (pid: 1524, threadinfo=d566c000 task=d58b8a00)

Stack: c011abe4 e0e21548 e0e215be e0e0f2d2 e0dff000 e0e0f3e8 000004d8 d566df8c
      c0143f2b e0dff000 e0e0f3e8 e0e26280 00000018 e0e26280 00000001 
00000001
      00000000 00000500 000004d8 000002d0 e09d7000 e0e26280 00000013 
00000000
Call Trace:
[<c011abe4>] module_finalize+0x94/0xa0
[<c0143f2b>] load_module+0x6cb/0x910
[<c014420b>] sys_init_module+0x9b/0x370
[<c016df05>] sys_read+0x45/0x60
[<c010a51b>] syscall_call+0x7/0xb

Code:  Bad EIP value.

You need to delete the __init from the definition of apply_alternatives().

2.5.68-mm2+e100 are now happy together -

Thanks to Andrew Morton for the fix:

delete the __init from the definition of apply_alternatives()

Joe