Hmm,
Last couple of times I watched my system boot up, it seemed not
to wait two seconds, but to go right on through(this on a Q650).
More of an annoyance/curiosity than anything else.  The only
side effect is that it lists my tape drive as empty, even though
a tape was left in.  Once the system is up, the "mt status"
confirms that a tape is mounted.  I guess the tape drive needs
a second or so to reseat the tape, etc. after a SCSI reset.

It looks like this stuff(tsleep) was reworked recently for the
multiprocessor modifications, so I'll try a new kernel and see if
it does the right thing...

Steve

Allen Briggs wrote:

> > I'm curious about a couple of lines in the boot messages below,
> > particularly the "waiting 2 seconds for devices to settle..." stuff.
> > What's going on there, something I can fix?
>
> Nothing bad--it's just giving the devices a chance to finish whatever
> initialization they might be in the middle of before it probes them.
> At least, that's my understanding.  Some devices do need it--most, I
> think, would probably be fine without it.
>
> > Also the savecore stuff near the end: is it safe to comment the
> > savecore line from /etc/default/rc.conf if I'm not experiencing crashes?
>
> Sure.  The savecore stuff shouldn't give you those errors, though, if
> your kernel and userland are in sync.
>
> > to the serial terminal.  Just out of curiosity, does anyone know what that
> > information is?
>
> You mean:
>
> > IOP base: 0x91d000
> > OLD cs0: 0x82
> > OLD cs1: 0x86
> > OLD cs2: 0x7 (i = 1)
>
> ?  This is debugging stuff from the (mostly disabled) IOP
> initialization.  It should be removed when someone gets around
> to working on that again.
>
> -allen