Re: rtVAX300 .. need help..

[Holm Tiffe <holm%freibergnet.de@localhost>]

Mouse wrote:

> > I think I have really simple and near stupid questions about the
> > architecture that may be is realized on that shitty ISA Board with
> > the CVAX on it.  Nobody tries to even recognize those questions.
> 
> Actually not true.  I do recognize the questions.  They're reasonable
> questions.  I've just kept silent because I have no answers for them.
> 
> > Again: What exactly are that MISC CPU internal registers?
> 
> I don't know.
> 
> > Are those [...]?
> 
> I don't know.
> 
> > [...], so I'm short on time.
> 
> This is a common complaint in this hobby.  I have 74 items on my "when
> I have the time" to-do list.  Most of them are not classiccmp things,
> but some are....

..as most of us and I really know that you have already invested much of
you time for that special thing.

> 
> > Next thing is, I'm not an natural english speaker, sometimes I have
> > problems to recognize how that's meant what one writes, sometimes I
> > have problems to express myself.
> 
> Your English is much, much better than my German.  It's true that
> misunderstandings can arise, but, for the most part, I don't think this
> is where the problem is.  My impression is that you have questions,
> perfectly reasonable questions, and are having trouble because you
> aren't getting answers.  But you probably aren't getting answers
> because nobody on the list _has_ the answers.  It's possible (albeit
> not especially likely in this case) that nobody now living has the
> answers.

Maybe. :-|
But even an answer "this may be related to.." or "really don't know.."
it's helpful, since this helps to supress the imagination of talking
against a wall..
> 
> > I was never in UK, never in the US and my english is coming entirely
> > from reading/writing Mails with people like you and from working on
> > Unix machines.
> 
> I am very impressed.  I do not think my German would be as good in the
> analogous case.

Thanks for the flowers :-)

It's not that much you that makes it difficult to me understand things..

> 
> > I got the first signs of life from that rtVAX in a relative short
> > time with the help from Mouse und yours, but with that little effort
> > that I could make the last days, the project "porting NetBSD to an
> > rtVAX300 board" will take for ever..
> 
> You have set yourself a relatively difficult task.  You have made
> progress faster than I would expect from someone who started out not
> even knowing any VAX machine - nor even assembly - language.  But it is
> still a difficult task.

Yes I know, but since nobody besides Dave has a similar hardware to play
with, this will be my part.

> 
> > In short: If I ask if there is an dedicated Interrupt controller in
> > the other VAXen or if it is normal that CPU REGs are mapped to
> > Address Space, then it isn't much help to read that none of the
> > machines are water cooled.
> 
> True.  Part of this may be that nobody knows the answers.  Part of it
> is probably the same topic drift that always happens on mailing lists.
> 
> I don't know the answers.  On the VAXen I know, there are some devices
> (the timer facility built around the NICR and ICCS IPRs, the console
> SLU on machines that use MTPR/MFPR console access, etc) that don't
> really have anything I could call an interrupt controller - or, to put
> it another way, the interrupt controller is built into the VAX IPL
> arbitration logic.  For other devices (Unibus, Qbus, etc), there is
> hardware for this; for example, how Qbus interrupts turn into VAX SCB
> vectors and processor priority levels is specific to the Qbus
> implementation in question.  I don't know how your rtVAX 300 does
> interrupt arbitration and mapping.  That information might be buried
> somewhere in the ROM code, in which case it might mean a lot of
> searching through the code to find it.  Or it might not, if the ROM
> code doesn't use interrupts.

Ther si something in the Hardware manual about this in a rtVAX300,
that wasn't te problem, I wanted some hints about other VAXen to just
understand the related NetBSD files.

It is interesting how DEC migrated from a "all known, all standard" system
on UNIBUS and QBUS where any register address and bit was known
to a wide varying family of machines where things are new all the time.
Exactly this is what makes me trouble since I don't know if a feature is
"standard" or an one-off occurrence.
> 
> I have trouble thinking where else you might find it, short of a lucky
> coincidence such as meeting, by chance, someone who happened to design
> the hardware in question.
> 
> I fear this is not very helpful. 

You are just wrong here.

> But you do appear to be breaking
> relatively new ground here, so, unless and until someone appears with
> the relevant hardware documentation from back when DEC made that
> hardware (or other information, such as someone who used to work with
> it), I don't have much to offer.  Not because I don't want to help, but
> because I have no help to give.

That's not true. Just keep talking with me, this helps a lot.

> 
> I may someday have the time to go through the ROM code in enough detail
> to figure that out.  But I don't have that kind of time right now any
> more than you do.
> 
> /~\ The ASCII                           Mouse
> \ / Ribbon Campaign
>  X  Against HTML              mouse%rodents-montreal.org@localhost
> / \ Email!         7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B

Don't think this will be neccessary, least not for now.


Regards,

Holm



-- 
      Technik Service u. Handel Tiffe, www.tsht.de, Holm Tiffe, 
     Freiberger Straße 42, 09600 Oberschöna, USt-Id: DE253710583
  www.tsht.de, info%tsht.de@localhost, Fax +49 3731 74200, Mobil: 0172 8790 741