> I got this shutdowning my machine. Maybe it is a pcmcia card services
> fault, but I think is interesting the oops generated by the
> exception handler while the first oops. Ramon Garcia spoke about
> this oopses some time ago...
>
> Oops: 0000
> CPU: 0
> EIP: 0010:[ds:unregister_pcmcia_driver+496/688]
> EFLAGS: 00010202
> eax: 02799b68 ebx: 00000001 ecx: 00000000 edx: 001a3308
> esi: 030422b0 edi: 00000001 ebp: 0019f26c esp: 0019f250
> ds: 0018 es: 0018 fs: 002b gs: 0018 ss: 0018
> Process swapper (pid: 0, process nr: 0, stackpage=0019d32c)
> Stack: 001170b1 00000001 00000001 ffffffff 00000001 00000001 f3080000 001bc8e0
> 0011c78b 00b45936 00000717 00000000 0019f2b0 0011611f 00b45936 0019f308
> 00000000 00000000 00000018 00000018 00b45936 0019f308 00000000 00009000
> Call Trace: [timer_bh+749/820] [do_bottom_half+59/96] [schedule+63/652] [hard_idle+75/92] [sys_idle+59/112] [system_call+85/124] [init+0/612]
> [start_kernel+409/420]
I've got pretty the same as above, with the same kernel and the
AGGRESIVE_PAGE_AGING patch aplied. The difference was that no call trace was
generated and EIP was 0. It also happened in the swapper. When the kernel
oopsed, the machine was busy compiling ghostscript 5.03. No log contained the
traceback. Oh, one more thing. Oops was preceded by hda (EIDE HDD) timeouts
when reading - the HDD is brand new and working correctly.