Re: NetBSD/vax kernel seems stable now

[Johnny Billquist <bqt%softjar.se@localhost>]

Hmm...

Still something very broken here.
I still crash at boot time, in readdisklabel()

I wonder if it might be when it's trying to figure out my second disk, which 
contains a VMS system, and thus don't have a proper disk label...?
Those of you who have succeded in booting now, do you have any unlabeled disks 
in your machines?

Grr. I can't seem to break into ddb with serial console on a VAX 4000/90. I'd 
really like to set a breakpoint right before the statement that causes the trap.

It's at readdisklabel+0x8d, and the code there is:

netbsd:readdisklabel+0x7f:      *clrl   -(sp)
netbsd:readdisklabel+0x81:      pushl   r6
netbsd:readdisklabel+0x83:      movab   brelse, r10
netbsd:readdisklabel+0x8a:      calls   $2, (r10)
netbsd:readdisklabel+0x8d:      movl    *0x28, r7
netbsd:readdisklabel+0x94:      pushl   40(r8)

It would be interesting to check r10 here...

        Johnny


Matt Thomas skrev:
> I fixed a thinko in the new softint handling code.
> I also revamped the mutex code.
>
> boot seems broken since it doesn't seem to be able to load kernels.
> (i'm using a boot i build a year ago or so and it works fine).

--
Johnny Billquist                  || "I'm on a bus
                                  ||  on a psychedelic trip
email: bqt%softjar.se@localhost             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol