Port-vax archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Floating CSRs [was Re: Address of second TMSCP controller]



On 2021-03-21 15:38, Mouse wrote:
[...], I suspect some would want to use the floating-CSR rules
simply bceause they were The Way It Was Done Back Then.
That's precisely the mindset at the museum, with DEC and other
systems.
Which is kindof weird, because pretty much noone, nowhere, that I
know of, did it that way back then.  [...] just grabbed any free CSR
address range [...]

That's what I would expect in an installation full of people who
understand Unibus addressing and the like.

It would not surprise me, though, if there were sites at places like
banks that had nobody on staff who understood anything and called their
four-hour-response field-service number whenever _anything_ went wrong.
It wouldn't surprise me if some such sites used floating-CSR rules not
because it produced a better system but simply because it was what DEC
officially recommended.  (Which it was, wasn't it?)

The only site I had anything to do with that was running a DEC OS on
Unibus/Qbus hardware was in academia, full of computer geeks.  We had a
device whose interface was multiple Unibus-form-factor cards that
required a custom-wired section of Unibus backplane, with a custom
driver for it.  The sort of place that used the term "field circus" and
told jokes like "How do you tell a DEC field service tech with a flat
tire?  He's the one swapping tires to see which one is flat.  How do
you tell a DEC field service tech who's out of gas?  He's the one
swapping tires to see which one is flat."  (This was back in the '80s,
when a field service tech who wasn't "he" was basically unheard of.)

We too did the just "grab a free range" sort of thing.

I never had any significant contact with the non-geeky sites; I have no
personal experience of what they did.  Does your experience include any
such?  In particular, does your "noone...did it that way" remark apply
to them too?  (I'm not challenging, just curious.  As I say, I have no
experience either way on the question myself.  But I would find it
mildly surprising if DEC had maintained these elaborate rules even
though nobody was paying any attention.)

Would banks ever change their hardware? I would usually guess not, which made it a pretty moot point around there. At initial installation, you usually had things setup according to recommendation. It also helped with initial install of software.

As I mentioned, RSX for example, do not try make any use of the floating address space convention. But there is one exception. When you do the initial SYSGEN from the baseline system, there is a tool that will probe your hardware according to that floating address space convention, and will provide default answers to a lot of SYSGEN questions.

So for initial installations, you usually were following the convention, and DEC had an interest in keeping it up to date from a documentation and installation point of view.

But once the system was set up, and running, any changes after that were usually done everywhere with the minimum amount of disruption.

And back then, even banks had a few geeks at hand, who dealt with all that dirty hardware. Kept them in a closet, so customers wouldn't see them. :-)

  Johnny

--
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


Home | Main Index | Thread Index | Old Index