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 17:28, Dave McGuire wrote:
On 3/21/21 9:36 AM, Johnny Billquist wrote:
On 2021-03-21 06:24, Dave McGuire wrote:
On March 20, 2021 5:09:00 PM Mouse <mouse%Rodents-Montreal.ORG@localhost> wrote:
Given the retrocomputing geeks who want to run systems as original as
feasible, even to the point of doing things like trying to find
original parts for power supplies and the like, 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.

  I've had the opposite experience.  Many years ago, in two jobs that had PDP-11s, and several after that had VAXen, they were all configured according to those rules.  All were installed by DEC, and most were under DEC maintenance.  One installation was under third-party maintenance.

  But perhaps the biggest reason why I enforce it on the DEC machines at LSSM is that I was taught that it was "the proper way" by my DEC mentor in the 1980s, and it stuck with me.

Well, I can't even remember if we stuck to that when I was working at DEC in the 80s, but I sortof have a feeling we didn't. :-)
The amount of effort if you do it was just too much.

  The only truly practical reason that I've heard in support of following those rules is that you can still get through your sysgen etc even if you've added a board and forgotten/lost your notes/etc where you put the CSR! ;)

Well, I can tell you that if you run a SYSGEN on RSX to reconfigure your system, you won't even have the option of probing using the floating space algorithm. The only time that is even an option is when you run the baseline system, which only happens at the original installation.

However, you have the saved answers from your previous SYSGEN still around usually, so what you did before was usually available. But usually, if you added a device in RSX, you didn't do any SYSGEN. You just built the device driver for the new device, added it to the system, and smiled (there is something to say for loadable device drivers...).

  In those installations, I can recall only one instance of hardware being added to an existing system (a tape drive was added to a MicroVAX 3600) so I can't say whether or not it was common to keep up with those rules in the field.  On my personal systems at home it was pretty common to throw in whatever new peripheral I was able to get my hands on, but in the production environments, we already had what we needed and just used it.

That on the other hand was pretty common. Many places didn't really change their hardware much over the years. So if it was configured according to the rules, it would still be so years later. Just because it wasn't changed anyway.

But if you wanted to just throw in a new controller, you didn't want to do a 8h SYSGEN under RSX, and then get everything else setup back the way you had it before, when there was the option of just compiling the driver, install it, and then you were done.


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