Re: DECstation 5000/200 timekeeping

>> So I conclude there's some pmax-specific breakage. There may well be
>> sme MI breakage, but the pmax behavior is clearly a regression.
>> (Mouse's report that pre-1.5 NetBSD has _much_ better timekeeping
>> (their emphasis) seems like good evidence.

I installed stock 1.4 and the clock refuses to misbehave.

Unfortunately, the _disk_ misbehaves.  Unless the hardware broke just
as I switched from 5.2 to 1.4, it's a software problem.

Fortunately, the network works, though until I find one of my pieces of
10base2 cable I need to connect it by resting one T connector on
another to get the shield connection and sticking a piece of wire into
the connectors for the centre connection.  I've been doing netboot, to
sidestep the disk troubles.  (It _is_, though, useful to be able to do
things like copy files around on the NFS server.)

I've been trying to bring it up to my 1.4T, but that's proving
difficult.  Every kernel I've built from my own source tree, even the
initial source tree that's the root of my change tree, wedges on boot.
I've been trying to debug it but haven't managed to determine much.

I added debugging printfs to init_main.c; they indicate that the exec
of init succeeds (returns -2, EJUSTRETURN) but as far as I can tell
init does not actually start.  So I added code to turn on ktracing,
basically copying much of sys_ktrace() inline in there, and now it's
wedging somewhere inside sys_execve() - I get a ktrace NAMI record for
/sbin/init, but tcpdump indicates that it's repeatedly trying to look
up "sbin", as if it were failing to receive the response packets.

It doesn't help that the keyboard on my principal house machine has
started to misbehave - the CONTROL key is flaky, sometimes not
registering even though it is physically pressed.  Occasionally I will
be typing something like a string of ^Fs and start getting f instead of
^F even though I haven't actually released the key.  I may have to swap
it out for another keyboard :( - fortunately, I have at last a few
spares.  Not directly related, but it makes it more difficult to work.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		[email protected]
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B