Matthew Dillon wrote:
> Ok, I think I have an idea as to what is going on. I think what is
> happening is that all the vm_map_entry structures are being eaten by
> other cpus and then the init process is started on a cpu that has yet
> to synchronize its reserve. That cpu then panics.
>
> If this is the problem then the solution is to reserve two vm_map_entries
> for each cpu so they can bootstrap themselves.
>
> Please try this patch, again using the original unmodified MAX_MAPENT.
>
> -Matt
>
>
Before I do that, I wanted to let you know that your previous hunch
about modules may be dead on. I've been able to boot up with nothing in
/boot/loader.conf
I am now going through my module list, adding them to loader.conf one at
a time and rebooting after each one so I can find out where it's
dying. So far I'm able to boot up with miibus, if_rl, if_fxp, if_lnc,
if_wi, and wlan loaded :-) That leaves agp, radeon, and bktr.
Added agp: fine.
Added radeon: fine.
Added bktr: Crash and burn.
As always, it's the last one you try :-)
Note: this is with the previous patch you posted. Do you still want me
to try with this most recent patch? If so, does it apply against CVS or
CVS + the previous patch? :-)
Adam