almost two months ago you wrote in reply to my problems with NetBSD 1.6
on a DecStation 5000/150:

> Can I suggest you repost your message to "port-pmax@netbsd.org".
> The port-pmax-maintainer email address only goes to two people,
> and you'll get a much broader audience if you sent your message
> to the main list.
> 
> I don't have a 5000/150 handy, but do have a 5000/50.  I'll see if
> everything works ok there.  Can you check that your machines does
> indeed still run ok with a 1.5.3 kernel?

I finally found the time to convert back to 1.5.3, and yes, it runs
nicely with the 1.5.3 generic kernel, just as it used to before I went to
1.6.
The timing problems with 1.6 release continue to exist.

Is there anything I could try to get 1.6 runnig without that SCSI timing
problems?



> 
> > why do I get those sd0 SCSI timeouts? As I said, with 1.5.3 the hardware ran
> > flawlessly. Now it crashes most of the time I boot up when entering single or
> > multi user mode.
> > 
> > After having loaded the generic kernel, just beforethe screen clears for the
> > above messages, I get:
> > 
> > mcclock loop count 264 too low for r4000
> > 
> > what does this mean? What kind of timing problems do I have?
> 
> The real-time-clock is used to calibrate the CPU speed.  If this
> calibration is incorrect, then it's possibly that some timing-related
> problems could occur.
> 
> > mcclock0: Cannot get 100 Hz clock; using 256 Hz
> > Again, there seem to be timing problems.
> > Anybody knows how to fix that? Is that a known issue?
> 
> It's not known to me.  I don't think that that part code really
> changed much; testing a 1.5.3 kernel should be helpful here too.
> 
> Cheers,
> Simon.
> --
> Simon Burge                                   <simonb@wasabisystems.com>
> NetBSD Development, Support and Service:   http://www.wasabisystems.com/
> 


-- 
Georg Schwarz    http://home.pages.de/~schwarz/
 geos@epost.de     +49 177 2437545