On Tue, Sep 04, 2007 at 06:43:42PM -0700, Paul Goyette wrote:
> 
> OK, I tried that and it didn't seem to help.
>

Bummer.
 
> 
> Not sure what process/thread 0.22 is since the keyboard is totally dead. 
> But based on the message, I'm pretty sure that IRQ 7 was the right one 
> to disable.
> 

Just an innocent victim I suspect, if you are in ddb you can try a ps
and see (well, if the keyboard worked you could).  When I last did
some digging on this I found that what was happening was when the
secondary processors were being spun up they were hit by a deluge of
interrupts which made the secondary processor enter the ISR
recursively until the kernel stack was exhausted causing a double
fault.  For me, reserving IRQ 7 made the problem go away... though, I
did also patch (iirc) the ehci driver to clear pending interrupts
after prising control of it away from the BIOS, I have no clue if that
helped or not.  From what I can see in the old thread it did not
appear to but maybe the combination of the patch and disabling irq 7
worked.  The patch is here:

http://mail-index.netbsd.org/current-users/2007/06/20/0009.html

The other thing that came up in that thread that I have forgotten
about is I updated the bios on my machine, that appeared to have some
impact too.  Any bios updates for your motherboard?

-- 
Brett Lymn