Re: Kernel behaving strangely

[Jukka Ruohonen <jruohonen%iki.fi@localhost>]

On Tue, Dec 28, 2010 at 12:10:48PM +0000, Sad Clouds wrote:
> Well I thought this might be something to do with Intel hyperthreading and
> ACPI, so I tried booting with ACPI turned off (boot netbsd -2).  This time
> I get 2 CPUs instead of 4 and everything seems to work,

The earlier (Intel) "MP specification" has been superceded by ACPI. All HT
siblings are only available with ACPI. The IO-APIC is typically enabled
only when you boot with ACPI. Long story short: you can not do without ACPI
in "modern" x86 systems.

> uvm_fault(0xffff8000493bbe70, 0x0, 1) -> e
> fatal page fault in supervisor mode
> trap type 6 code 0 rip ffffffff80723a8f cs 8 rflags 10293 cr2 0 cpl 8 rsp 
> f000498dbb40
> kernel: page fault trap, code=0
> Stopped in pid 629.1 (halt) at netbsd:acpi_enter_sleep_state+0x2f: movq0 
> (%rdi),%rdi
>
> This shouldn't happen should it?

Of course not. I think this is fixed in HEAD.

As for the other problems, interrupt issues would be my first guess. 

- Jukka.