> From: "Harry Schreurs" <HLS@oce.nl>
> Date:          Fri, 11 Nov 1994 17:35:31 GMT +0100

> Gordon,
> 
> I would like to know that the "segmentation fault" still exist
> in the kernel of Oct 25!

Yes, sorry, I know.  I've been frustrated too.  It's been only fairly
recently that I've been able to put in enough time to make progress
on the VM/pmap problems.  The code is complicated, so it takes long
sessions to "get my head" into the code enough to make progress.

> This is no complaint. I tried very hard to understand the underlying
> pmap code.  But without proper documentation this is very hard.

It has taken quite a bit of study time for me as well...

> >   db> call sun3_rom_halt
> > 
> I didn't know that this call was available.

You can call anything you wan from DDB, but the sun3_rom_* functions
were designed to be convenient for use in both DDB and the kernel.

> I will volunteer to test this version if yout want!
> I have 2 Sun 3/50's to test this on. One has 4Mb of memory
> and the other one has 8 Mb.

I'll be sure to make lots of noise here when I have something
ready for testing.  There is a 3/50 on it's way to me also.

> Where can I get the source and/or compiled ready to run kernel?
I'll check-in what I've got as soon as it is better than what is
contained in the 1.0 release.  I've made quite a few changes from
that release at this point, so it would not be convenient to try
to send out parts of those changes.  Anyway, in another week or so
the SUP server on lamp should be providing "current" again so the
changes I'm making can be distributed through that channel.

If you (or others) want to work on things, there's always the
code for various drivers, vme bus support, etc...

Thanks for your patience,
Gordon Ross